Vous êtes sur la page 1sur 387
______________________________________ Recent advances in security technology Proceedings of the 2006 RNSA Security Technology Conference Canberra, 2006

______________________________________

Recent advances in security technology

Proceedings of the 2006 RNSA Security Technology Conference Canberra, 2006

Editors: Priyan Mendis, Jospeh Lai and Ed Dawson

______________________________________

Recent advances in security technology

Recent advances in security technology The RNSA is an Australian Government initiative. The Research Network for
Recent advances in security technology The RNSA is an Australian Government initiative. The Research Network for

The RNSA is an Australian Government initiative.

The Research Network for a Secure Australia (RNSA) is a multidisciplinary collaboration established to strengthen Australia’s research capacity for protecting critical infrastructure (CIP) from natural or human caused disasters including terrorist acts. The RNSA facilitates a knowledge-sharing network for research organisations, government and the private sector to develop research tools and methods to mitigate emerging safety and security issues relating to critical infrastructure. World-leaders with extensive national and international linkages in relevant scientific, engineering and technological research will lead this collaboration. The RNSA also organises various activities to foster research collaboration and nurture young investigators.

Participants are encouraged to join the RNSA. Membership of the RNSA is open to Australian and international researchers, industry, government and others professionally involved in CIP Research. Information on joining is at www.secureaustralia.org.

Convenor:

A/Prof Priyan Mendis, Head of the Advanced Protective Technology for Engineering Structures Group, The University of Melbourne

Node Leader:

Prof Ed Dawson, Queensland University of Technology

Node Leader:

Prof Joseph Lai, UNSW@ADFA

Research Program Manager:

Dr Tuan Ngo, The University of Melbourne

Technical & Administrative Support Officer:

Anant Gupta, The University of Melbourne

Financial & Management Support Officer:

Maggie Burke, The University of Melbourne

Outreach Manager:

Athol Yates

Published by the Australian Homeland Security Research Centre. The Centre undertakes independent, evidence-based analysis of domestic security issues. Australian Homeland Security Research Centre Tel 02 6161 5143, Fax 02 6161 5144, PO Box 295, Curtin ACT 2605 AIIA Building, Level 1, 32 Thesiger Cct, Deakin ACT info@homelandsecurity.org.au, www.homelandsecurity.org.au

ISBN 0-9757873-4-9, 978-0-9757873-4-2 © 2006, Australian Homeland Security Research Centre, and the authors.

All rights reserved. Other than brief extracts, no part of this publication may be produced in any form without the written consent of the publisher. The Publisher makes no representation or warranty regarding the accuracy, timeliness, suitability or any other aspect of the information contained in this book and cannot accept any legal responsibility or liability for any errors or omissions that may be made.

Foreword

Recent advances in security technology

The 2006 Science, Engineering and Technology Summit was the second formal conference organised by the Research Network for a Secure Australia (RNSA) funded by the Australian Research Council. The Summit has become an annual event bringing together both researchers and practitioners in the fields relating to the national research priority area entitled Safeguarding Australia.

It provided a forum for the exchange of ideas and research findings between core groups or individuals interested in security technology.

The plenary session focused on a panel discussion on the challenges and opportunities in establishing joint industry, research institution and government projects. Mr. Mike Rothery, Assistant Secretary, CIP, Attorney-General’s Department, Dr. Richard Davis, Head of NSST Unit, Dept. of Prime-Minister and Cabinet and Mr. Jason Brown, ADI Limited participated in this forum. There was also a presentation from Mr. Michael Jerks and Dr. Greg Scott on the Critical Infrastructure Protection Modelling and Analysis Program (CIPMA).

The conference had both refereed paper and technical update presenations, and two special security technology seminars on Global Navigation Satellite Systems (GNSS) and on Security & Technology research at National ICT Australia (NICTA).

The papers in this volume were those accepted into the refereed streams. Each paper was subjected to a rigorous review process conducted by at least two experts in the appropriate field. The authors were requested to revise the papers according to reviewer’s comments.

The editors would like to thank all of the reviewers for their assistance in maintaining the high quality of papers.

Finally the editorial committee wishes to thank the publisher of these proceedings, the Australian Homeland Security Research Centre and its Executive Director Athol Yates.

A/Prof. Priyan Mendis Prof. Joseph Lai Prof. Ed Dawson

Recent advances in security technology

Reviews

The editors sssss

Table of contents - DRAFT

Recent advances in security technology

Recent advances in security technology

1

Development of the RNSA transport strategy for Australia:

considerations for passengers and freight

M.A.P. Taylor

University of South Australia, Adelaide, Australia

Abstract

This paper provides some background to and a description of the emerging transport strategy that is being developed in RNSA. It outlines the research dissemination program that is proposed under this strategy and how this relates to established international collaboration in transport research. The connections between the RNSA transport strategy and contemporary issues in transport research such as network planning and critical infrastructure, sustainability and the use of advanced (ITS) technologies are introduced.

Biography

The author is the Professor of Transport Planning and Director of the Transport Systems Centre at UniSA. His expertise is in transport modelling, intelligent transport systems and transport network reliability. He received his BE, MEngSc and PhD degrees at Monash University, where he was an academic on two separate occasions. He has also worked as a traffic engineer, as a consultant with the Road Transport Research Programme of OECD, and as a research scientist in CSIRO. Taylor is a Fellow of the Institution of Engineers, Australia, the Chartered Institute of Logistics and Transport, and the Institute of Transportation Engineers.

Introduction

Recent advances in security technology

Safety and security of transport systems offer a complex, multidimensional problem for transport planners, and one that is accentuated because of the fundamental needs of a modern society for personal mobility and access to goods and services.

The transport security group in RNSA is developing a transport strategy for the research network, which recognises the needs for security considerations in transport planning and identifies the dimensions pertaining to transport security. A particular emphasis in our initial work is a concern about the identification of critical elements of transport infrastructure and the development of plans to deal with failure or degradation of infrastructure. These plans need to cover both the short term – how to deal with an emergency – and the long term – what are the societal impacts of infrastructure loss or degraded performance. Impacts on the movements of people and of goods have to be considered. There has been intense international research on transport network reliability over the past decade (e.g. see Bell and Cassir (2000), Bell and Iida (2003) and Murray and Grubesic (2006)), and this research has a strong and natural influence on the RNSA transport strategy.

Given the nature of a research network as opposed to a research program, the draft transport strategy for RNSA focuses on the dissemination of research questions and research results rather than the formulation of a research agenda per se, although it is to be expected that an agenda will emerge – or an existing agenda will be modified – as a result of the activities and discussions in the research network.

The draft transport strategy

The draft RNSA transport strategy was initiated at a meeting held at the University of Melbourne in July 2006, when a group of transport researchers in RNSA decided to pursue a transport strategy under the umbrella of the research network. The group comprised Assoc Prof Kim Hassall (University of Melbourne), Dr Russell Thompson (University of Melbourne), Professor Marcus Wigan (Napier University/University of Melbourne) and the author – who is greatly indebted to his colleagues for their ongoing contributions to the development of the strategy.

The overall aim of the strategy is to develop broader and more intense academic and practical involvement for transport activities in the RNSA, and to raise the profile of transport security, risk and reliability in that network and in the transport planning community.

Under the auspices of RNSA, this aim can be pursued

by setting up appropriately targeted meetings and workshops on topics including

o

transit

o

supply chain

Recent advances in security technology

o

network reliability

o

social impacts

o

associated infrastructure (ports, airports, large manufacturers, etc)

o

utilities and government

by aiming for operational follow up actions, investigations and projects.

In consequence the RNSA transport strategy seeks to pursue an immediate set of transport-related activities in RNSA, as follows:

ensure that the September 2006 program of workshops on ‘Transport network

reliability considering risk, security and emergency issues’ is effectively delivered contribute to the RNSA Security Technology conference in Canberra on 21

September 2006 continue the development of international collaboration in research on

transport reliability and security, drawing on the established collaboration with Imperial College in the UK and the established international research network on transportation network reliability provided by the International Symposia on Transportation Network Reliability (INSTR) and its International Advisory Committee follow up the workshop program with the London based workshop, to be held

at Imperial College (with Australian input from the University of South Australia, the University of Melbourne, and the University of Sydney). It is expected that this workshop will be held in May-June 2007 set up European linked follow up activities to expand on the inputs and

lessons, possibly using FEAST continue to address critical transport infrastructure issues (led by the Transport

Systems Centre at the University of South Australia) address supply chain issues (led by the Transport Security Research Group at

the University of Melbourne) target conferences, especially in Asia and Europe. The 7th International Conference of the Eastern Asia Society for Transportation Studies (EASTS07) (Dalian, PRC, September 2007 and the 3rd International Symposium on Transportation Network Reliability (INSTR07) (Delft, The Netherlands, 18-22 July 2007) are of particular interest in this regard. Locally, the annual Australasian Transport Research Forum provides a suitable venue.

The draft strategy is a work in progress and a substantial evolution of this program is to be expected.

A first initiative

The first major transport initiative of RNSA is a series of three workshops on ‘Transport Network Reliability considering Risk, Security and Emergency Issues’ to be held in Melbourne, Adelaide and Sydney in September 2006. The workshops feature presentations by a number of invited eminent speakers from the UK, The Netherlands

Recent advances in security technology

and Australia. The workshop contents include examination and reliability analysis of urban transport networks, security implications for public transport systems, modelling and analysis of pedestrian flows in areas of high levels of pedestrian activity, planning for evacuations of both city centres (e.g. in the event of an attack) and suburban areas (e.g. because of bush fires). One or more local speakers representing police, emergency services or a transport planning agency in each of the three cities are also included.

The workshops are designed to cover a spectrum of security and safety related issues for transport systems, including natural, manmade and malicious events, and short and long term concerns involving transport systems operations and management, and tactical and strategic planning for transport systems and infrastructure.

A rationale

Wigan and Clarke (2006) provided a basis for the consideration of transport security issues in transport planning, in their paper at the ‘1st Workshop on the Social Implications of National Security’ held in Wollongong in May 2006. They emphasised that while the transport sector is an obvious focal point for surveillance measures aimed at combating threats of terrorism, it is also a complex environment that highlights the social impacts of surveillance. In particular, they stressed the importance of trust to the public acceptance of intrusive measures such as surveillance and screening systems, the dangers of the dissipation of that trust, and the potential threats to freedoms in our society such as the right of freedom of anonymous movement.

Wigan and Clarke thus point to two major areas of interest in contemporary transport research and in the practice of transport planning: mobility and accessibility and the monitoring, collection and use of data on individual travel behaviour.

Accessibility is identified as major determinant in both travel demand and on the development of human settlement systems, and therefore the interaction between transport and land use systems (and environmental impacts) (Wegener, 1996).

Real time monitoring of the movement of vehicles and individuals is an important facet of the operation of modern ‘Intelligent Transport Systems’ (ITS), which rely on information and communications technology to improve both the efficiency and security of transport systems operation, e.g. through electronic toll collection for vehicles or contactless smart cards for public transport passengers. A major social issue in the use of these technologies, which can provide unique identification of individual vehicles or travellers, is the scope for the potential wider usage of the data and concomitant concerns about protection of individual privacy. This has led, for instance, to the development of an Australian Standard (AS4721-2000) (Standards Australia, 2000) which expressly limits the use of such information solely to the primary purpose for which it was collected.

Recent advances in security technology

Two other areas of research in transport are of importance in considerations of transport security. These are transport network reliability leading to the identification of critical transport infrastructure and social, economic and environment sustainability of transport systems. Considerations are needed for both passengers and freight, as recognised in the RNSA draft transport strategy.

Critical transport infrastructure

The concepts of transport network reliability and transport accessibility are being applied in current research on the development of methods for identifying critical elements of transport infrastructure, using the notion of transport vulnerability. Vulnerability may be defined in terms of the potential social and economic impacts of failure or degradation of some part of a transport system. A useful metric for estimating these impacts is through the change in accessibility that may result from network failure or degradation. The concept of vulnerability and the use of accessibility indices to assess vulnerability impacts are described in Taylor, Sekhar and D’Este (1996). Choice of suitable accessibility metrics can allow vulnerability analysis at a variety of planning levels, from the large scale strategic level through regional studies to the local level.

Sustainability

Contemporary transport planning is dominated by the concern about sustainability of transport systems, especially in urban areas. Global warming, greenhouse gas emissions, environmental degradation and dissipation of scarce liquid fossil fuel resources are combining to stimulate major research initiatives in transport.

A useful definition of sustainability in transport is that suggested by Minken (1999), which is based on the work of Chichilinsky (1996) and Heal (1998). According to Minken, one of the two defining characteristics of sustainability as an objective is that it includes both the welfare of present day society and the society of the very distant future. The second defining characteristic is that sustainability implies conservation of natural resources. Thus natural resources should be valued not only as something that may be consumed (in production or consumption) but also as stocks that benefit us even when not being consumed. We depend on some basic qualities of our surrounding ecosystems for our quality of life and indeed for our continued existence. On this basis, a sustainable urban transport and land use system is one which:

provides access to goods and services in an efficient way for all inhabitants of an

urban area protects the environment, cultural heritage and ecosystems for the present

generation does not endanger the opportunities for future generations to reach at least the same welfare level as that of the present generation, including the welfare derived from the natural environment and cultural heritage.

Recent advances in security technology

Transport and urban planning for sustainability tends to address two major issues: the form and structure of our cities (notably the density and mix of land use developments) and the improved usage of ‘environmentally friendly transport modes’, which include walking and cycling and public transport (May and Taylor, 2002). The factors affecting choice of transport mode by passengers have been the subject of long term research (e.g. Hensher, 2001) and amongst these factors is known to be personal security (along with cost, convenience and comfort, amongst others).

Freight

Considerations of freight transport, especially urban freight and goods distribution, and sustainability are also important in transport research. In fact, given that the area of freight transport research has been largely neglected, this is the burgeoning area for future research. One particular direction for freight transport research with connotations for both sustainability and security is ‘city logistics’, which attempts to solve a multiple objective optimisation problem relating to the efficiency and certainty of urban freight distribution with minimum environmental impact (Taniguchi, Thompson, Yamada and Van Duin, 2001).

In line with movements in transport planning practice, transport research is starting to address the wider issue for freight of supply chain management. The security of supply chains then becomes a major issue for research and investigation.

Conclusions

Transport security is now an issue in the public eye, almost on a daily basis as we are told about potential catastrophic events hopefully averted, or from time to time bear witness to actual events.

In fact, transport safety and security issues provide a spectrum of circumstances from the minor and perhaps mundane to the serious and even world shattering. The ‘iceberg analogy’ fits well with this spectrum.

Transport research needs to be concerned about security issues, for both the immediate operations of our transport systems and their roles in the economic and social wellbeing of our society, and in the longer term planning of transport systems, services and infrastructure cognisant of wider societal concerns such as sustainability, conservation of scarce resources, and social equity and justice.

There is a substantial role for a research network such as RNSA to play in the encouragement of transport research and in the dissemination of research outcomes. The draft transport strategy for RNSA introduced in this paper provides an initial step towards fulfilling this role. Much more work is required to develop the strategy and comment and inputs from interested researchers and practitioners are essential.

Recent advances in security technology

Acknowledgements

The contributions of Assoc Prof Kim Hassall, Dr Russell Thompson and Professor Marcus Wigan to the development of the RNSA transport strategy are substantial and ongoing, and this paper could not have been written without their support and encouragement. In addition, Professor Peter Stopher of the Institute of Transport and Logistics Studies at the University of Sydney has had a substantial influence on the directions of the strategy. His support is also acknowledged.

References

Bell, M G H & Cassir, C (eds) (2000). Reliability of Transport Networks. Baldcock, Herts: Research Studies Press. Bell, M G H & Iida, Y (eds) (2003). The Network Reliability of Transport. Oxford: Elsevier-Pergamon. Chichilinsky, G (1996). An axiomatic approach to sustainable development. Social Choice and Welfare 13 (2), pp.231-257. Heal, G (1998). Valuing the Future: Economic Theory and Sustainability. New York: Columbia University Press. Hensher, D A (ed) (2001). Travel Behaviour Research – The Leading Edge. Oxford: Elsevier-Pergamon. May, A D & Taylor, M A P (2002). KonSULT – developing an international knowledgebase on urban transport policy instruments. Papers of the Australasian Transport Research Forum 25. October, Canberra. CD-ROM, Paper no 40. Canberra: Bureau of Transport and Regional Economics. Minken, H (1999). A sustainability objective function for local transport policy evaluation. In H Meersman, E, Van De Voorde and W Winkelmans(eds), World Transport Research. Selected Proceedings from the 8th World Conference on Transport Research. Volume 4: Transport Policy. Oxford: Elsevier-Pergamon, pp.269-279. Murray, A T & Grubesic, T H (eds) (2006). Reliability and Vulnerability in Critical Infrastructure: A Quantitative Geographic Perspective. New York: Springer-Verlag (in press). Standards Australia (2000). Australian Standard: Personal privacy practices for the electronic tolling industry. AS 4721-2000, Sydney: Standards Australia. Taniguchi, E, Thompson, R G, Yamada, T & Van Duin, R (2001). City Logistics: Network Modelling and Intelligent Transport Systems. Oxford: Elsevier. Taylor, M A P, Sekhar, S V C & D’Este, G M (2006). Application of accessibility based methods for vulnerability analysis of strategic road networks. Networks and Spatial Economics 6, pp.267-291. Wegener, M (1996). Reduction of CO2 emissions of transport by reorganisation of urban activities. In J R Roy & Y Hayashi (eds), Transport, Land-Use and the Environment. Kluwer Academic Publishers:

Dordrecht, pp.103-124. Wigan, M R and Clarke, R (2006). Social impacts of transport surveillance. In Katrina Michael & M G Michael (eds), The Social Implications of Information Security Measures on Citizens and Business. Proceedings of the 1st Workshop on the Social Implications of National Security, Wollongong, 29 May 2006. Wollongong: University of Wollongong, Centre for eBusiness Application Research (CeBAR).

Recent advances in security technology

2

The Success and Failure Factors for Incident Management Systems

Rob Bailey B.Bus, B.Acc, Dip.Crim.

Dependable Property Services Pty Ltd, Melbourne, Victoria, Australia On Line Risk Management (ORM) Pty Ltd

Abstract

In my experience, many organisations are devoid of at least some of the factors required for effective incident reporting. A number of critical matters are often overlooked and, when issues arise, it is often difficult and sometimes impossible to repair the damage. An attitude of ambivalence seems to prevail, that is, once an incident report is submitted, little if anything is done other than registering and filing it…until an issue arises. While many organisations have some of the procedures and protocols in place for effective incident management, unless they have all the components, all they really have is information.

The purpose of this presentation is to provide organisations with a ‘template’ from which they can implement a proven method of effective incident management. Organisations that implement the total package are guaranteed a positive return in terms of reduced premiums, reduced legal fees and greatly improved corporate and industry reputation.

Biography

Nine of Rob’s 29 years with Victoria Police were spent managing major fraud investigations. He project managed the implementation of the Police & Emergency Services Communications Centres, and supervised the Business Continuity Plan for both Victorian and New Zealand Emergency Services Communications. Rob has extensive experience in investigating mission-critical incidents, including those that

Recent advances in security technology

involved Coronial hearings. He retired as a Chief Inspector, and now divides his time between Dependable Property Services (providing facility management) and On Line Risk Management (incident management services) in the capacity of part- owner/Director.

Rob completed Vicpol Detective and Advanced Detective Training courses, was the Director of Studies for the National Commercial Crime Course, received six Chief Commissioner Commendations, and worked with the United Nations in Cyprus during the Turkish Invasion. He has assessed over 50,000 incidents, and been involved in security and risk management in retail shopping centres for 28 years.

Introduction

Effective incident reporting relies on a number of major factors. In my experience, many organisations are de void of several of these factors, which include:

  • 1. An effective incident report.

  • 2. Effective training of those compiling the report and involved in the incident.

  • 3. An efficient and dynamic Incident Reporting ‘system’.

  • 4. Compliance, transparency and auditability.

  • 5. Site risk management, including the registers.

  • 6. A proactive approach.

One of the most critical mistakes made by organisations is in using the number of incidents as a key success indicator in terms of the reduction in hazards and the likelihood of an ‘incident’. All this does is inspire ground staff to fail to report incidents! The upshot of this is, if you don’t know about it, you can’t treat it. The correct measure of success is the number of liability claims lodged. Organisations should demand the reporting of the most seemingly innocuous incidents, as it is often these events that cause the surprises and indeed provide useful information regarding many operational issues!

One client (in the oil industry) offered incentives for staff to report incidents, and the number of reported incidents increased by 400%. This provided significant strategic information to assist the company to make sound decisions.

When called to review a client’s existing incident management/reporting system, almost without exception the attitude is one of ambivalence – simply file it and wait. My analogy to this inaction is akin to parachuting: ‘You’re stuffed until something goes right!’

The Incident Report

Recent advances in security technology

Many incident reports are mandated either by an insurance broker, the insurer or the organisation’s risk managers. The most complete reports originate from those who actually work in the organisation’s risk environment. The reason for this is that insurers are only interested in the likelihood and value of a claim, while internal risk managers have broader information requirements.

In my experience, we have been able to encourage the brokers and insurers to adopt a more appropriate format in order to obtain other information for the benefit of the organisation. For example, I believe it is imperative to capture mobile phone numbers from all third parties and witnesses. People change their addresses, telephone numbers, even their names, but rarely their mobile number. This provides the most reliable search parameter for duplicate or multiple claimants…and we have ample evidence that this occurs. Any incident management system should be able to match duplicate and/or suspicious incident reports.

Training

Typically, security personnel are relied upon to compile incident reports in the retail sector. However, there are many other industries, for example the oil and petroleum industry, where untrained, casual employees who are manning the ‘shop’ are called upon to compile reports. Indeed, in some cases, it is virtually impossible to train everyone! How than can this be managed?

In one example, this was overcome by providing sole operators with the option to either phone through the details or fax the incident report. In fact, the telephone option was encouraged, as a trained operator could then record the details provided by the sole operator, using a blank incident report as a template. As time passed, many of the sole operators became accustomed to the requirements, thus being able to accurately complete the report themselves prior to faxing it. Initially, about 65% telephoned the details through, whereas now, only 10% use this method, and there is a very low error rate!

The adage really is, ‘If you don’t know about it, you can’t do anything about it’. Therefore, it is essential that training is synonymous with a regime of mandatory reporting. This can be achieved by encouragement and reward, or as a KPI - both work effectively.

In appropriate environs, training must include all facets of incident reporting:

thoroughness, accuracy, and initial investigation. This should include statements, photographs and much more. However, even if all that is reported is the bare minimum, an effective pro-active system will provide the stimulus to all other criteria required.

Recent advances in security technology

The System

There are a number of incident management procedures, protocols and ‘systems’ that can facilitate the effective management of incidents. Again, based on my experience, typically an incident report is prepared and forwarded to interested parties, who put it in a file and wait for something to go wrong, then act! A number of companies offer ‘claims management’ services. Whilst I acknowledge that these firms may do a very good job once a matter has become a claim, I strongly advocate that what is really needed is pro-active action before an incident becomes a claim.

Much of the expense and attendant corporate reputation impact can be avoided if matters are effectively handled as soon as practicable after they occur. This should include:

Automated circulation to all stakeholders.

Immediate assessment of the incident and notification of assessment to

stakeholders. Direct correlation with site risk register to determine if any new hazards have

been identified by the incident, and which should be included in the risk register. Identify if a duplicate incident or suspicious matter has occurred.

Initiate any further follow-up required, eg. information, investigation,

statements etc. Direct contact with third party – this is the most crucial ingredient to

success!** Further assessment of the incident and notification to stakeholders.

Ongoing management of file, including any actions required from the third

party call. Compliance with privacy requirements.

Create a funnel effect as incidents are closed, so that focus is placed on ‘open files’. ** History and factual evidence shows that a professional and timely third party call to the person involved in an incident achieves the following:

  • (a) A feeling that someone cares.

  • (b) An accurate indication of the demeanour, attitude and intentions of the person.

  • (c) Opportunity to close-out the matter with a minimum of time and expense.

  • (d) Changing attitude by repeat or suspicious claimants.

Statistics reveal that of the 740-plus incidents occurring in the past 5 years in major shopping centres where this procedure is followed, only one small claim occurred.

Privacy & Auditability

Recent advances in security technology

The Privacy Act (2001), as amended, requires certain details to be given to people from whom information is taken. My research shows that many organisations either don’t provide these details, or do a poor job of it. It is essential that any system contemplated is privacy-compliant.

Additionally, the system should be totally auditable, transparent and accountable. It is indisputable that organisations that have a professional bullet-proof incident management system are far more likely to enjoy a significantly better result when there is a major claim.

Site Risk Register

Many organisations incorrectly marry the Site Risk Register to the OH&S Risk Register. Every site must have a professionally completed ‘living document’ Site Risk Register. This register should be directly linked to all new incidents to enable immediate hazard identification and treatment.

In many instances, effective compilation and maintenance of the risk register will ameliorate the contractor organisation’s exposure.

Proactive Incident Management

Whilst iteration of an earlier point, it is most important that organisations develop a totally proactive approach to incident management. This starts with processes and protocols at the site(s), to the implementation of a system, and handling of each matter individually. This can be managed in-house or outsourced. In either case, properly facilitated, it will be more than cost-effective.

Examples (3)

How things can go horribly wrong…Just a small cut on the finger

This is a situation where a shopper sustained a small cut on his finger, which was dressed by a trader and a cleaner. Two days later he was on life support .On the surface, such incidents appear innocuous and more often than not do not draw any immediate action. The fact that an effective system was in place meant that this incident was identified and managed.

Recent advances in security technology

Multiple fraudulent claims…How a systems approach can negate fraudulent claims

A woman (referred to as ‘Woman A’) slipped/tripped on an escalator. The incident was witnessed by her daughter-in-law’s sister. Two weeks later, the daughter-in-law slipped on an escalator, which was witnessed by Woman A!! The system matched this as a duplicate entry, and the third party calls supported the fact that both incidents were fictitious.

A ballerina slipped twice in a car wash, injuring both knees, the incident being witnessed by her husband. Attention to detail during constant contact with this lady revealed a very suspicious situation.

Effective use of statistics…Making the right judgment, based on fact

The use of an efficient incident management system revealed that a significant number of incidents (slips/trips) were occurring between 11am and 2pm on most Tuesdays near the shopping centre food court. Investigation revealed that it was a as a result of elderly people collecting their pension at the nearby Bank.

What to do? Many contractors would put a further cleaner on duty to obviate the problem. Wrong! The implementation/relocation of a security guard eliminated the occurrences. The presence of the security guard made people more careful, and his role in identifying hazards and relaying this information to the cleaners whilst monitoring the spill/hazard was successful.

Conclusion

Notwithstanding that recent legislation has achieved some limiting of public liability claims, lawyers are already finding ways around making outrageous claims. Further, the legislative changes have not had any impact on minor (gratuitous) claims, legal expenses and organisation reputation. In every circumstance, firms want to provide clean, safe and healthy environs for the public, with a minimum fuss and interference. In reality, this can only be achieved by ensuring that every component of the incident management process is implemented and professionally managed.

Recent advances in security technology

2

The Risk Transaction Cycle, And Its Role In National Security

Richard A. Nunes-Vaz & Svetoslav Gaidow

Defence Science & Technology Organisation, Adelaide, Australia.

Abstract

National security relies on the effectiveness of many intra- and inter-agency relationships and interfaces. Observations from the Melbourne 2006 Commonwealth Games identified that security agencies routinely ‘trade’ risk without, necessarily, having an explicit understanding that this is occurring. Using the concept of a transaction from electronic banking, a framework has been developed which facilitates risk exchange between stakeholders, without loss or corruption. Each stakeholder expresses risk in different terms, and each transaction typically generates at least one critical product or decision. Explicit recognition of the risk transaction process allows a number of tools and devices to be employed that clarify agency roles, make risk exchange decisions traceable and auditable, and make the requirements on outputs transparent.

Biography

RICK NUNES-VAZ: is a senior analyst working in both the Defence and civilian sectors of national security. Since joining DSTO in 2000 and prior to 2005, he led areas of capability and methodology development in support of Defence experimentation. Rick has a physics background (Imperial College, London) completing his PhD in oceanography in 1982 (University of Wales). He also holds postgraduate qualifications in arts and engineering. He held academic positions (in oceanography) at Flinders University of South Australia and University of New South Wales until 2000 and has an extensive publication record in a number of disciplines.

SVETOSLAV GAIDOW gained Masters (Probability & Statistics, 1980), and PhD degrees (Mathematical Sciences, 1986) from the University of Sofia and the Mathematics Institute of the Bulgarian Academy of Sciences. Following positions with

Recent advances in security technology

the Universities of Rousse, Plovdiv and the Centre for Mathematics and Computer Science in Amsterdam (with interests in stochastic differential equations, optimal control and game theory) Svet moved to an academic position with RMIT, in Melbourne. Svet joined DSTO in 2002 and, following a 2-year posting to Army’s Land Warfare Development Centre, he moved to Land Operations Division in Adelaide where he applies risk management principles to Defence problems.

  • 1 Introduction

In the loose confederation of agencies that characterise the national security architectures of most western nations, interoperability is a long-term goal. Yet success, expressed in risk management terms, depends critically on the effectiveness of many intra- and inter-agency risk negotiation processes.

Our observations of the communication between Victoria Police (VICPOL) and the Australian Defence Force (ADF) during the Melbourne 2006 Commonwealth Games (M2006) revealed the generic properties of these risk exchanges, and allowed us to distil its essential character (Nunes-Vaz et al, 2006).

This paper describes what we call the risk transaction cycle. It is a generic description of risk exchange that occurs when the responsibilities for a risk reduction objective are shared between different agencies, or different departments of the same agency. While the trading of risk management responsibilities is ubiquitous in the national security arena, the lack of common language and processes clouds its generic character, and its actors may fail to appreciate their roles and exactly what needs to be achieved. The risk transaction cycle is intended to provide a conceptual template to clarify roles and responsibilities, provide traceability and auditability, and define critical outputs.

  • 2 Terminology

In order to minimise ambiguity we have developed a model to guide the use of particular terms, and to describe the relationships between those terms (figure 1). It was based upon, and is consistent with, the Risk Management Standard (AS/NZS 4360: 2004), the Defence Security Manual (Commonwealth of Australia, 2005) and the National Counter-Terrorism Handbook (Commonwealth of Australia, 2005).

Recent advances in security technology

Recent advances in security technology Figure 1. Terms and their relationships in the national security risk

Figure 1. Terms and their relationships in the national security risk management context

In addition to the terms identified in Figure 1, we also define:

Assurance

the likelihood of achieving objectives (AS/NZS 4360: 2004)

Owner

the ‘problem-owner’ and all of its decision-making delegates. The

problem-

owner is ultimately responsible for security. This was VICPOL in

M2006.

Partner

an agency, and all of its decision-making delegates, that is tasked by the owner to support risk reduction activities. The ADF in M2006 was a partner.

Strategic ‘big picture’, longer-term view, including ramifications across all sectors of concern

Operational

all current and near-term areas of activity and influence

Tactical

localised and immediate activity

To strike a balance between generality and clarity, we illustrate points with aspects of the security-related interactions between VICPOL and the ADF during M2006,

Recent advances in security technology

although our use does not necessarily reflect anything that actually occurred during

M2006.

  • 3 The Nature of Risk Exchange

Each agency speaks of risk in different terms. The problem owner might describe its concerns as dignitary protection or maintenance of reputation. The partner must convert these risk objectives into response activity (search) objectives, which might be expressed in the form of effort, locations and timings.

We found it useful to develop an analogy for these exchanges, as a form of currency conversion. The language, or currency, used by each agency to describe its risk concerns and its risk reduction activities is specific to its perspective. So the strategic risk concern of avoiding casualties must be converted effectively into a tactical risk reduction activity of searching this room rather than that.

Extending the analogy, we also adopted the concept of a transaction. The inter-agency negotiation is a risk transaction, in the same sense that occurs in electronic banking. A transaction is a process that guarantees the integrity of a commodity (in this case risk) through an exchange process (negotiation). The risk transaction that gave the ADF responsibility for searching venues must ensure that the communication of initial risk, risk concerns and risk reduction objectives occurs without loss or corruption.

It is possible to describe all steps in the risk reduction process (in this case associated with search operations) as a series of risk transactions. The series should form a cycle, because ultimately the risk reduction achieved must be compared with the strategic risk reduction objective (that is in the same currency) in order to decide whether further risk treatment is necessary.

  • 4 The Risk Transaction Cycle

Figure 2 shows the risk transaction cycle that applied during the venue search operations of M2006. It has six steps (risk exchanges) between each pair of double- boxes (the ‘search’ box explicitly situates the search activity in relation to the overall risk reduction process). Each step of the cycle is discussed in turn below. We deliberately develop the currency analogy by identifying the nature of the conversion or exchange process, and the rate of exchange, or how the currencies are related. The output of each transaction is, in our experience, one or more critical decisions, although we have not resolved whether this is actually a rule.

Recent advances in security technology

Recent advances in security technology Figure 2. The Risk Transaction Cycle as it applied to M2006

Figure 2. The Risk Transaction Cycle as it applied to M2006

Step 1 – From Assurance Objective to Search Objectives

Initial Currency:

Converted Currency:

Exchange Process:

Rate of Exchange:

Decision(s):

Assurance Objective Search Objectives Stakeholder negotiation Opportunities for harm, to their potential physical expression (eg, MCG, Games Village, Transport Systems…) The need for searches, and raising the search force

For the problem owner, risk is expressed as an assurance requirement. It is couched in terms of maintenance of reputation, adherence to event schedules etc. Risk assessment confirms the need for risk reduction, and consideration of treatment options leads to the decision that it will be achieved (partially, if not wholly) by venue searches.

The risk transaction converts the component of risk reduction required from searches, to search objectives. For an effective (lossless) transaction, VICPOL must express and prioritise their (risk-related) concerns clearly and objectively, and the ADF must convert those concerns to their potential physical expression within the specified search context.

Recent advances in security technology

Step 2 – From Search Objectives to Search Priorities

Initial Currency:

Search objectives

Converted Currency:

Search priorities

Exchange Process:

Search appraisal (risk assessment) of venues in the

Rate of Exchange:

context of the event’s security overlay Opportunities for harm to its spatial distribution in

Decision(s):

specified zones The allocation and movement of searchers and

equipment.

Step 2 was an intra-agency transaction in the case of M2006, taking place between the ADF’s strategic acceptance of its role, and the operational units tasked to carry it out.

The ADF response was to deploy search advisors to conduct venue appraisals. Their methods codify their risk assessment to disaggregate the risk into three zones: high risk (Red); medium risk (Amber) and low risk (Green): called a RAG analysis. The ADF trains its searchers to more than one level of competency, so completing the transaction also involved identifying how the different skill levels (and equipment) were to be used.

Step 3 – From Search Priorities to Elements of Risk

Initial Currency:

Converted Currency:

Exchange Process:

Rate of Exchange:

Decision(s):

Search priorities Elements of residual risk (limitations) Search: elimination, by inspection, of elements of risk Extent of elimination of risk (in terms of risk reduction per unit of effort) by the application of search capability Remediation of risk elements as they arise

This risk transaction was an intra-agency process, involving conversion from the operational (ADF) unit to the tactical tasking of individuals to ensure effective conduct of search activities.

Other than in the simplest of search contexts, unsearched elements will remain that represent residual risk, called limitations. An example of a limitation might be a locked room, or a locked cabinet inside a room, where the keyholders are unavailable.

In reality many remediation decisions are made during the conduct of searches when limitations are first identified, so decisions would be made by stakeholders (eg, ADF, VICPOL and the cabinet owner’s delegate) whether to be more forceful and cause damage. The remediation decisions apply to the constraints of time (is more time required), access (permission to inspect) and intrusiveness (physical capability to inspect).

Recent advances in security technology

Step 4 – From Elements of Risk to Search Outcomes

Initial Currency:

Converted Currency:

Exchange Process:

Rate of Exchange:

Decision(s):

Elements of residual risk (limitations) Search outcomes Expert appraisal (by partner, specifically the search advisor) Contribution of limitations to original search objectives The acceptability (or otherwise) of search outcomes

The ADF’s search advisor is an expert role. During the conduct of a search the search advisor must monitor its progress (see Figure 3) in terms of the rate of risk reduction (elimination of elements of risk) and the aggregate effect of remaining limitations.

Step 5 – From Search Outcomes to Assurance Achieved

Initial Currency:

Converted Currency:

Exchange Process:

Rate of Exchange:

Decision(s):

Search outcomes Assurance achieved Expert appraisal (by owner) Residual risk, and its meaning with respect to the likelihood of achieving the owner’s objective (assurance) Further (local) risk treatment

Recent advances in security technology Step 4 – From Elements of Risk to Search Outcomes Initial

Figure 3. The aggregate risk picture before and after commencement of search activities

This was an inter-agency transaction (at the operational level) in which the ADF communicated (and formally handed over) the outcomes of the search to VICPOL.

Recent advances in security technology

More generally, the information must be articulated in sufficient clarity for the owner to be able to assess remaining potential impacts on high-level objectives (eg, ensuring that no one is killed, and the events run on time).

The handover must empower the owner to make ‘local’ decisions. An example of a local decision might be the decision to divert a dignitary walkthrough to maintain adequate distance from an identified limitation.

Step 6 – From Assurance Achieved to Assurance Required

Initial Currency:

Converted Currency:

Exchange Process:

Rate of Exchange:

Decision(s):

Assurance achieved Assurance required Expert appraisal / risk assessment (by owner) No conversion required – currencies are the same Further (global) risk treatment

This was a strategic intra-agency transaction conducted entirely within the confines of VICPOL.

Comparing the risk reduction objectives to the risk reduction achieved (from all treatment options activated, not just search operations) drives decisions on whether, and what, further risk treatment is required.

  • 5. Traceability and Auditability

Effective risk transactions should be traceable and auditable, that is clarity exists not only in the minds of the delegates conducting the transactions, but clarity is also manifest in the documents that express their deliberations and decisions.

Close examination of each step of the risk transaction cycle suggests conceptual mechanisms that facilitate clarity. An example is shown in Figure 4. Presentation of other such devices is beyond the scope of this paper.

Recent advances in security technology More generally, the information must be artic ulated in sufficient clarity

Figure 4. An example of clear, prioritized expression of strategic risk concerns.

Recent advances in security technology

Figure 4 is intended to be the product of the owner’s risk analysis prior to engaging with a partner (to assist Step 1). It identifies a number of levels of consequence, and specific prioritised concerns under those levels.

For example, it indicates that this (fictitious) owner considers the “death of royalty” to be of more concern than the need to abandon the event (though both are considered “catastrophic”). We stress that this example is entirely fictitious and does not reflect any real information or discussion related to M2006.

6.

Summary

This paper describes a generic device, called the risk transaction cycle, that is intended to support the discussions and decisions of agencies involved in risk management activities, particularly with respect to national security. It is intended to support both intra-agency and inter-agency negotiations, and promotes clear articulation, traceability and auditability of the security process.

Acknowledgements

The authors are grateful to the following persons for opportunities to discuss and develop the ideas contained in this paper: Dr Fred Bowden, LTCOL Michael Buck, LTCOL Adam Boyd, COL Mark Richards, LTCOL Glenn Stockton and MAJ Paul Foura.

References

Commonwealth of Australia, Attorney-General’s Department, 2005, The National Counter-Terrorism

Handbook (unclassified parts), Robert Garran Offices, National Circuit, Barton ACT 2600. Commonwealth of Australia, Department of Defence, 2005, Defence Security Manual,

http://intranet.defence.gov.au/dsa/dsm/index.html

Nunes-Vaz, R.A., Bowden, F., Buck, M., Ellis-Steinborner, S. & Hobbs, W., 2006, Towards Search Benchmarks for Australia’s National Security, DSTO-CR-06-0121, Defence Science & Technology Organisation, Edinburgh, South Australia, 5111.

Standards Australia/Standards New Zealand, 2004, AS/NZS 4360:2004 Risk Management, Standards Australia International, Sydney, NSW 2001

Recent advances in security technology

4

A Framework for Estimating Economic Costs and Human Casualties of Hazards

L. Marquez, A. Ahmed & R. Kusumo

CSIRO Mathematical and Information Sciences, Australia

TO BE RECEIVED : 12 Pages

Recent advances in security technology

5

An Update on Full Scale Blast Tests in Woomera, 2006

R. Lumantarna, A. Gupta, P. Mendis & T. Ngo

University of Melbourne, Australia

J. Ramsay

Ballistic Consultant, Australia

TO BE RECEIVED : 8 Pages

Recent advances in security technology

6

Polymer Reinforced CRC For Improved Performance Under Close-In Detonations

  • B. Riisgaard

NIRAS Consulting Engineers and Planner, Denmark Centre for Protective Structures and Materials, Technical University of Denmark and APTES, The University of Melbourne (Visiting Researcher)

  • A. Gupta, P. Mendis, T. Ngo

Advanced Protective Technology for Engineering Structures Dept .of Civil & Environmental Engineering, The University of Melbourne

  • A. Elkjaer

Composhield A/S, Denmark

Abstract

Compact Reinforced Composite (CRC) is a high-strength cement-based composite, which has an enormous flexural and energy-absorbing capacity due to its closely- spaced high-strength steel reinforcement and a high-strength cement-based fibre DSP matrix. The material has been used in various constructions, including as protection for explosion hazards. In connection with the explosive impact, the fraction of shear reinforcement needed to obtain full flexural capacity is controlled by the stand-off distance. For close-in detonations, a high fraction of shock reinforcement is needed to obtain full flexural capacity without breaching. This paper introduces an efficient method for implementing high fractions of polymer shock reinforcement into a CRC element.

Biography

Recent advances in security technology

Since 2000, Benjamin Riisgaard has been working as a developer and consultant regarding blast and ballistic protection. He has experience with the design of protective solutions for embassies and critical infrastructure sites, and has developed various military ballistic protective systems. In 2004, Benjamin Riisgaard established The Centre for Protective Structures and Materials at the Technical University of Denmark. In 2005, he was employed at NIRAS Consulting Engineers and Planners, Denmark, and in the same year started an industrial PhD project concerning the modelling of blast protective High-Performance Fibre-Reinforced Concrete structures. Currently, Benjamin is working as a visiting researcher at APTES.

1.

Introduction

Advanced high performance concrete composites and their response to blast loading is an area of continuing interest within the field of civil engineering. Polymer reinforcement in concrete composites and the elements for protective application include glass, aramid (Kevlar) or polyvinyl alcohol fibre bundles. HPFRC (High-Performance Fibre-Reinforced Concrete) of any kind refers to fibre- reinforced cement-based materials that are developed for specific applications where strength, toughness, ductility, and energy absorption are fundamental properties. These properties in HPFRC are obtained by using large quantities of super plasticizers, high volumes of micro-silica, a low water-cement ratio, high fractions of small discrete fibres, and the absence of any coarse aggregate.

A HPFRC like FDSP (Fibre-reinforced Densified Small Particle systems) used in CRC (Compact Reinforced Composite) applications holds compressive strength in the range of 150 to 400 MPa (Bindiganavile et al., 2002).

This paper introduces an efficient method for implementing high fractions of polymer shock reinforcement into an element.

The term shock reinforcement refers to reinforcement in the out-of-plane direction that is implemented to avoid the shock-initiated disintegration of a structural element.

The term fibre bundles refers to a large number of parallel fibres of the same length that have been spun into a cord.

The composite presented in this paper is patented by Composhield A/S Denmark. The numerical modelling of the work presented in this paper is being carried out as a part of ongoing project with the APTES group.

Recent advances in security technology

  • 2. The effect of a close-in detonation

Detonation and explosion, terms which are commonly used synonymously, produce a shock front travelling faster than the speed of sound. Blast wave propagation can be treated mathematically on the basis of thermodynamics by solving the conservation equations analytically (Kinney and Graham, 1985). A contact or a close-in detonation is characterized by the absence of, or limited, stand-off distance from the explosive charge to, for example, a structural element, and it generates shock waves in the element of up to 30 times the speed of sound; pressures of up to 20 Gpa; and strain rates of up to 10 8 s -1 , depending on the material and charge size.

Recent advances in security technology 2. The effect of a close-in detonation Detonation and explosion, terms

Figure 1 - Contact detonation on a concrete plate, regions of different stress states and failure situations

A common effect of close-in detonations on concrete structures is spalling, which leads to disintegration of the structural element. Spalling is caused by the free surface reflection of the shock wave induced in the structural element by a high-pressure air blast, and occurs whenever the dynamic tensile rupture strength of the structure is exceeded. Although it is a complex process, reasonable analytical spall estimates can be obtained for concrete by assuming elastic material behaviour and instantaneous spall information. Specifically, spall thicknesses and velocities for both the normal and oblique incidence of the shock wave on the back face of the structural element can be calculated.

  • 3. Compact Reinforced Composite

Compact Reinforced Composite (CRC) is a Fibre-reinforced Densified Small Particle system (FDSP), combined with a close-spaced, high-strength, longitudinal flexural rebar arrangement.

Recent advances in security technology

Recent advances in security technology Figure 2 – Principe of 50 mm CRC panel The addition

Figure 2 – Principe of 50 mm CRC panel

The addition of small discrete fibres to the DSP matrix makes it more homogeneous and transforms it from a brittle to a ductile material. Because the fibres are randomly oriented and very closely spaced throughout the matrix, they are more effective than conventional reinforcement for bridging across cracks, and the presence of the fibres also provides some post-cracking ductility. The ultimate shear strength increases with an increase in fibre volume fraction, and in some cases full flexural capacity can be reached in elements without stirrups just by using fibres. (Khuntia et al., 1999)

  • 4. Compact Reinforced Composite improved for close- in detonation

CRC improved for close-in detonation is a Fibre-reinforced Densified Small Particle system (FDSP), combined with a close-spaced high-strength longitudinal flexural rebar arrangement laced together in the out-of-plane direction using polymer lacing to avoid shock initiated disintegration of the structural element.

Recent advances in security technology Figure 2 – Principe of 50 mm CRC panel The addition

Figure 3 – Principle of 50 mm CRC deck improved for close-in detonation

Aramid (Kevlar) was developed in the late sixties as a new class of polymers—para- oriented polyamides (aramids)—possessing internally rigid molecular chains in an

Recent advances in security technology

extended confirmation. Like glass and carbon fibres, the tensile stress-strain curve of aramids is almost linear up to the point of failure. The details are shown in Table 1.

The combination of low density with high-strength and elastic modulus gives aramid the highest specific tensile strength of any material and a reasonably high specific modulus, even when compared with carbon fibre (Nanni, 1992).

Specific

Tensile strength

Elastic modulus

Elongation

 

[%]

gravity 1.39 – 1.45

[MPa (ksi)] 2640 (383) – 3040 (441)

[GPa (10 3 ksi)] 75.5 (10.9) – 127.5

2.0 – 4.2

(18.5)

Table 1 - Characteristics for aramids

  • 5. Experimental work

Two panels of 50 mm thickness, which were tested, are included in the paper. In Table 2 the dimensions of the panels, charge size, and stand-off are shown. The panels were reinforced by 1800 MPa orthogonal longitudinal flexural high-strength steel rebars on both faces Φ 5 mm, and a centre-to-centre spacing of 10 mm as shown in Figure 7. The cover layer was 3 mm. The panels were simply supported by a steel frame with a free inner dimension of 500 x 500mm. The panels were tested under identical support conditions, but under different boundary conditions, to gain sufficient anchor length.

Panel

Dimension

Charge PETN (eqv.

Distance*

Distance**

[mm]

TNT) [kg]

[mm]

[mm]

1

50x600x600

  • 1.0 (1.3)

40

85

2

50x1200x1200

  • 3.5 (4.5)

40

126

Table 2 – Dimensions, charge size and stand off * Distance surface of charge to surface deck ** Distance (eqv. TNT) centre of charge to surface panel

Recent advances in security technology extended confirmation. Like glass and carbon fibres, the tensile stress-strain curve

Figure 4 – Experimental setup for 50x1200x1200 panel – 3.5kg of PETN HE

Recent advances in security technology

The tests presented in this paper were based on a matrix design that offers sufficient workability and maximum design strength and ductility. The formula used for these mix proportions was:

where

V

b

=

1

V

agg

V

f

w

/

c

3150

1000

+

1

V

w

1

= −

V

agg

VV

−−

b

f

V b is the binder fraction

V agg is the aggregate fraction w/c is water-cement ratio V f is the percentage of fibres V w is the water fraction For the tests presented in this paper the following design was chosen:

Aggregate

Binder

Water/cement

Fiber

[vol%]

[vol%]

[ratio]

[vol%]

0.3

0.42

0.18

4

China Bauxite

Binder

Fiber, Bekaert

Comp.

size [mm]

Product name

Length, diameter [mm]

strength

[MPa]

0-1

Densit Primer (ready mix)

6/0.16

160-170

Table 3 – Polymer CRC design matrix used in the tests presented in this paper

  • 6. Experimental results

Recent advances in security technology The tests presented in this paper were based on a matrix

(a) Figure 5. a) 50x600x600mm – 1 kg PETN / 4 cm

(b) b) 50x1200x1200mm – 3.5 kg PETN / 4 cm

Recent advances in security technology

Observations (Panel 1): 50 x 600 x 600 – 1 kg PETN / 4 cm

Plastic deflection was measured to 18 mm. No, or only little, anchor failure was seen. The cover layer on the front side (warm side) suffered from scabbing. No lacing was cut over and all longitudinal rebars were intact. The cover layer on the rear side (cold side) spalled, but the lacing was intact, as shown in Figure 6.

(a) Rear Side (b) Front Side
(a) Rear Side
(b) Front Side

Figure 6. 50x600x600mm – 1 kg PETN / 4 cm - a) Rear side b) Front side

Observations (Panel 2): 50 x 1200 x 1200 – 3.5 kg PETN / 4 cm

Plastic deflection was measured to 60 mm. No, or only little, anchor failure was seen. The cover layer on front side (warm side) suffered from severe scabbing. Locally, some lacing was cut over, but all longitudinal rebars were intact. The cover layer on the rear side (cold side) spalled, but the lacing was intact as shown in Figure 7.

(a) Rear Side (b) Front Side
(a) Rear Side
(b) Front Side

Figure 7. 50x1200x1200mm – 3.5 kg PETN / 4 cm - a) Rear side b) Front side

7.

Conclusion

Recent advances in security technology

The main purpose of this test was to demonstrate the potential of this modified CRC composite subjected to close-in detonation. Demonstrations were made on two polymer reinforced CRC panels improved for close-in detonation with dimensions as follows: Panel 1: 50x600x600mm and Panel 2: 50x1200x1200mm. The panels were subjected to 1.0kg and 3.5kg of PETN, respectively, at a stand-off distance of 4 cm surface-to-surface for both panels. The results showed no breaching of or any damage to the reinforcing bars. For Panel 1, plastic deflection was measured as 18 mm. Very little anchor failure was seen. The cover layer on the front side (warm side) suffered from scabbing (cratering). No lacing was cut over and all longitudinal rebars were intact. The cover layer on the rear side (cold side) spalled, but the lacing was intact. For Panel 2, plastic deflection was measured as 60 mm. Very little anchor failure was seen. The cover layer on the front side (warm side) suffered from severe scabbing (cratering). Locally, some lacing was cut over, but all longitudinal rebars were intact. The cover layer on the rear side (cold side) spalled, but the lacing was intact. The numerical modelling of the work presented in this paper is being carried out as part of an ongoing project with the APTES group.

References

Bindiganavile, V., Banthia, N., and Aarup, B., 2002, Impact response of ultra-high- strength fiber-reinforced cement composite: Aci Materials Journal, v. 99, p. 543-

548.

Khuntia, Madhusudan, Stojadinovic, Bozidar, and Goel, Subhash C. Shear strength of normal and high-strength fiber reinforced concrete beams without stirrups. ACI Structural Journal 96[2], 282-289. 1999. Kinney, G. F. and Graham, K. J. Explosive shocks in air. Springerverlag . 1985. Ref Type: Generic Nanni, Antonio. Properties of aramid-fiber reinforced concrete and SIFCON. Journal of Materials in Civil Engineering 4[1], 1-15. 1992.

Recent advances in security technology

7

An engineering-level tool for predicting airblast loads behind blast perimeter barriers and their effectiveness

A.M. Remennikov

University of Wollongong, Australia

T.A. Rose

Defence College of Management and Technology, UK

Abstract

Urban planners often consider the use of barrier walls as a means of protecting structures and critical infrastructure facilities from terrorist vehicle bombs. Depending upon their height and structural integrity, barrier walls may also provide significant shielding from the airblast produced by the bomb. Several methods exist for the prediction of this shielding effect. Unfortunately, these methods tend to be rather inaccurate, due primarily to the complexity of the airblast/barrier wall interaction.

An improved methodology based on neural network model is proposed to predict peak pressure and impulse behind barrier walls of various heights for a wide range of explosive-to-wall and wall-to-structure distances. This paper provides an overview of the experimental and computational techniques involved in the development of the blast effects database and presents details of a new PC-based tool for rapid evaluation of the blast loads behind a blast perimeter barrier.

Biography

Alex Remennikov is a Senior Lecturer in Structural Engineering at the University of Wollongong. He received his doctoral degree from the Kiev National University of

Recent advances in security technology

Construction and Architecture in 1993. In 1994-1997, he undertook post-doctoral training at the University of Canterbury, New Zealand in the area of designing structures under severe earthquake conditions. His interests are in the area of analysis and design of structures subjected to abnormal loading such as impact and blast. He has been involved in a number of projects related to blast protection of the Australian critical infrastructure facilities.

Dr Timothy Rose studied Civil Engineering at University College London and for the last twenty years has worked in the area of blast and ballistic loading of structures at Cranfield University in the United Kingdom. In recent years, his efforts have been concentrated on the numerical simulation of blast from high explosive events in built-up areas. He is also the author of the popular Air3d computer code for numerical simulation of blast wave-structure interaction.

Introduction

Terrorism is evolving as a major threat to critical infrastructure and personnel throughout the world. In recent years car bomb attacks in which an explosive laden vehicle is detonated as close as possible to a target building have become a weapon of choice by some terrorist groups. The most common response to mitigate this threat has been to increase standoff distance (bollards, anti-ram walls, etc.) and retrofit existing buildings with blast protection measures. Increasing vehicle standoff distance could significantly increase real estate and construction costs and could be very difficult to achieve in a dense urban environment. One cost-effective alternative is the installation of perimeter walls for blast effects reduction that would decrease the need for standoff and building blast retrofit requirements. However, only limited information is available on the blast loading environment behind a blast wall (Mayor & Flanders 1990, DAHSCWEMAN 1998).

This paper reviews some available experimental data in this area of research and presents a new approach to developing an engineering-level tool for predicting the effectiveness of blast perimeter walls. Such a tool could provide the opportunity to conduct a simplified and rapid analysis of the advantages of a blast protective barrier without the need for lengthy numerical simulations and feasibility studies.

Improved predictive tools for the assessment of the blast environment behind a protective perimeter wall would allow urban site planners to account for any benefit from the wall in protecting structures and other critical facilities from vehicle bombs. The proposed engineering tool can also be employed for assessment of explosive storage and munition manufacturing facilities, where reinforced concrete dividing walls are used as shields for personnel protection and as physical barriers between explosive production steps.

Recent advances in security technology

Use of perimeter barriers for blast mitigation

Blast loads in simple geometries can be predicted using empirical or semi-empirical methods (Remennikov 2003). These can be used to calculate blast wave parameters for hemispherical surface charges or spherical free air explosive charges to predict blast effects on isolated structures and structural components.

Events of the recent years have demonstrated that the most common source of unplanned explosions were terrorist devices in an urban environment. In complex geometries, consisting of multiple buildings or other obstacles capable of modifying blast propagation, the blast wave behaviour can only be predicted from first principles using such numerical tools as SHAMRC (Crepeau 1999), AUTODYN (Century Dynamics 2005), Air3D (Rose 2003), and others. Such tools solve the governing fluid dynamics equations and can be used to simulate three-dimensional blast wave propagation including multiple reflections, rarefaction and diffraction. In addition, Computational Fluid Dynamics (CFD) techniques can capture such key effects as blast focussing due to the level of confinement and blast shielding by barriers or other buildings.

Recent advances in security technology Use of perimeter barriers for blast mitigation Blast loads in simple

Figure 1. Propagation of shock waves behind blast barrier

A blast perimeter barrier serves to reflect back some portion of the explosive blast energy and thereby reduces the blast environment behind the wall. The effectiveness of the wall in suppressing the blast environment depends on the parameters of the wall and the explosive charge. When placed across the path of the blast wave, the barrier may reflect a portion of the incident wave and cause the remaining part of the blast to diffract over the wall, as illustrated in Figure 1. The resulting diffracted pressure will be significantly reduced, and the distance behind the wall where the blast environment is significantly reduced may characterise the effectiveness of the barrier.

Recent advances in security technology

The US Army Security Engineering Manual (TM5-583-3 1986) contains a methodology for calculating the reduction in pressures and impulses behind a blast wall, but the manual is restricted for official use only and has a very limited availability to the non-military engineering community. The formulations in the manual are based upon small-scale tests conducted at US Army Engineer Waterways Experiment Station (WES) in the 1980s (Dove et al. 1989). These formulations were never validated with full-scale experiments.

The Protective Structures Automated Design System (PSADS) (DAHSCWEMAN 1998) provides some guidance on the effects of barriers and blast walls. PSADS was developed by the US Army Protective Design Centre to automate the procedures for blast-resistant design of structures. It provides methods for the design of protective structures in such areas as airblast, fragmentation, blast loads on structures, dynamic response of structures, as well as a suite of support computer codes. PSADS provides design charts for the peak pressures and impulses behind blast walls for two scaled wall heights (0.436 m/kg1/3 and 0.872 m/kg1/3). The design charts are based on experimental studies that had a limited range of burst positions, barrier heights, and measurement locations. As a result, the use of the design charts as a predictive engineering tool has very limited application.

Recent advances in security technology The US Army Security Engineering Manual (TM5-583-3 1986) contains a methodology

Figure 2. Geometric configuration of blast wall analysis scenario

More recent research results for blast walls were reported by Humphreys & Piepenburg (2001) and Bogosian & Shi (2002) on the same series of explosive tests. In this 35-test series, the experiments included a number of barrier configurations, some of them rigid concrete walls, and others being frangible barrier designs. One of the major findings from this test series was that thin, frangible barriers could be almost as effective as their more massive rigid counterparts.

Recent advances in security technology

To predict the reduced level of pressures and impulses behind a blast wall, a simple methodology is included in standard manuals (e.g. TM5-583) and uses the geometric configuration shown in Figure 2. The perimeter wall, of height H and thickness K, is located a distance L2 from the explosive charge W kg TNT and a distance L3 from the face of the building. The standoff distance between the charge and the building, if there were no perimeter wall, would be a distance L1. The height at which reflected pressure and impulse are measured on the building is T. The same geometric configuration applies when side-on blast resultants in free air are required behind a blast wall for personnel protection outside the building.

The methodology involves determination of adjustment factors, for pressure (AF P ) and impulse (AF i ), which are defined as follows:

AF

P

=

P

with barrier

I

with barrier

P

no barrier

I

no barrier

and

AF

i

=

(1)

The adjustment factors represent the ratio of the pressure and impulse behind the blast wall to its original value without the blast wall (i.e. assuming a standoff distance L1). Thus, an adjustment factor of 0.3 represents a 70 percent reduction in pressure and impulse behind a wall compared to a no-wall configuration.

The adjustment factors are obtained from the blast load adjustment curves in the manuals (TM5-583-3 1986) as a function of parameters W, L1, L2, H, and T. To use the curves, one needs first to generate the value of pressure and impulse that would be expected at the location of interest if there were no perimeter wall. Typically, the no- barrier blast loads can be estimated using simplified analytical relationships or design charts (TM5-1300 1990) as a function of scaled distance L1. They may also be evaluated using the computer programs such as ConWep or ATBlast (available from the US General Services Administration website (GSA 2006)) that automate those calculations. As a second step, an adjustment factor is then obtained from the curves for a given perimeter wall, bomb size, and building placement. The no-wall pressure and impulse values are then multiplied by the adjustment factors to produce the values behind the blast barriers. In cases where blast loading is required for the area of interest on the building face, the adjusted pressure and impulse can be averaged over this area of interest to provide loading for analysis and design.

Rose et al. (1995) carried out a programme of research in which detailed measurements of the blast environment were made behind a 1/10th scale vertical barrier. For the set of parameters studied in that work, it was concluded that in the region between three and six wall heights behind a blast wall and over a vertical distance of three wall heights above the ground, pressures and impulses may be reduced to no more than 60% and 80% respectively of those without a wall. The data sets of measured pressures and impulses from the experimental programme by Rose et al. (1995) are used in this study to develop a neural network-based system for prediction of the blast environment behind a blast wall.

Recent advances in security technology

The blast environment behind the wall may be further suppressed by employing a canopy near the top of the wall on the loaded side, as shown in PSADS (DAHSCWEMAN 1998). The use of a canopy assumes that blast wave would be focused in a direction away from the protected target. The canopy could be blown away by the forces of the explosion, but it would remain in place long enough to reduce the magnitude of blast wave diffracting over the wall. The effectiveness of a canopy is not considered further in this paper.

Predicting the blast environment behind a vertical wall barrier using Artificial Neural Networks (ANNs)

A blast wall is a barrier placed between the location of a possible explosive threat and that which it is protecting (personnel, inhabited building, critical infrastructure facility, etc.). The problem of predicting blast effects behind the vertical barrier can be approached by conducting (a) small-scale experiments (Rose et al. 1995); (b) full-scale experiments (Humphreys & Piepenburg 2001, Bogosian & Shi 2002); or (c) Computational Fluid Dynamics (CFD) simulations (Rice et al. 2000). The first two approaches are the most expensive and time consuming, and typically limited in a number of blast wall scenarios and measurement locations. The use of 3-D CFD blast propagation calculations can generate data sets for a large area of space behind the wall, for a wide range of wall heights, and at practically unlimited numbers of target points to give a detailed picture of the blast environment behind a barrier. Despite the relatively simple geometry of a blast wall scenario, it has been demonstrated (Rose 2001) that the blast wave-blast wall interaction is particularly difficult to analyse using CFD techniques. This difficulty is caused by the fact that the process of a blast wave diffracting over the wall is highly dependent on the pressure and duration (wavelength) which the blast possesses as it passes the top of the wall. In numerical simulations of a blast wall scenario, this process would primarily depend on the mesh discretisation. In real-life scenarios, the blast wall thickness would be just a small fraction of the computational domain dimensions which can result in the wall model being represented by only one or two computational cells across its width. Such mesh discretisation would be unrepresentative of the continuum, and one should expect degradation of the simulated blast effects behind the wall, depending on other problem parameters such as standoff distance, distance behind the wall, charge weight, and so on. Therefore, the experimental data sets containing values of pressures and impulses behind a blast wall are a more reliable source of data that can be used to calibrate the numerically based results. In this paper, the experimental data sets for a relatively wide range of scaled wall heights and a large area of scaled space behind the wall are employed to train and validate the neural networks. Three-dimensional CFD calculations can be used to extend the database by considering a number of additional blast wall scenarios. These will be presented in the subsequent publications by the authors. The goal of the present study is to develop a fast-running tool for predicting the blast environment behind a vertical barrier. This is accomplished by training an artificial neural network to approximate the overpressures and impulses from the experimental

Recent advances in security technology

programme by Rose et al. (1995). The present study uses the blast scenario nomenclature as shown in Table 1. From Table 1, it can be seen that every blast wall scenario can be completely defined by the following independent scaled using cube root scaling (Mays & Smith 1995) parameters:

Parameter

Units: [m/kg 1/3 ]

  • 1. scaled standoff distance

R/W 1/3

  • 2. scaled charge elevation

Z/W 1/3

  • 3. scaled distance behind the wall

R/W 1/3

  • 4. scaled height of the wall

H/W 1/3

  • 5. scaled elevation of point of interest

h/W 1/3

Table 1

These five parameters uniquely define the blast wall scenario. Therefore, these five parameters will be used as inputs to train and validate the neural network using the experimental data sets.

The problem of characterising the blast environment behind a vertical barrier on the basis of existing experimental data is essentially a prediction (interpolation) problem. Since artificial neural networks are proving to be an effective tool for predicting values of blast loads (Remennikov & Mendis 2006), the basic idea in a neural network based approach is to train a neural network with patterns of the blast wall scenario parameters describing the spatial distribution of blast wave parameters in the area behind the wall. This implies that each pattern represents the unique values of peak pressure and peak scaled impulse at each of the monitoring locations behind a blast wall with height H due to detonation of an explosive charge, W, at a particular location described by values of R and h. Therefore, the patterns of the quantities describing the blast environment are used as inputs and the peak pressure and peak scaled impulse as outputs to train the neural network.

The training of a neural network with appropriate data containing the information about the cause and effect is a key requirement of a neural network approach. This means that the first step is to establish the training set which can be used to train a network in a way that the network can predict the blast effects within reasonable accuracy of the experimental results.

In order to verify how well a trained network has learned the training cases, the trained network is tested by subjecting it to the training sets. The important generalization capability of a neural network for predicting blast wave parameters is tested by subjecting the trained network to data not included in the training sets (the so-called validating sets). How well a trained network is to generalize depends on the adequacy of the selected network architecture (number of hidden nodes, number of hidden layers, and so on) and the information on the blast load environment included in the training sets.

Preparation of the data sets

Recent advances in security technology

The steps for developing ANN models typically include determination of model input and output variables; division of the available data into training, testing, and validation subsets; determination of the optimal network architecture; and model validation. In this study, the most comprehensive set of experimental data available to the authors (Rose 1995) is used to develop the data subsets. This provides the opportunity to consider a relatively wide range of scaled wall heights (h/W 1/3 = 0.5 to 1.0 m/kg 1/3 ) and a large area of scaled space (R/W 1/3 = 0.35 to 5.0 m/kg 1/3 ) behind the wall.

The experimental programme was designed to include blast walls of three different heights. The charges were Demex plastic explosive, and they were initiated by No.8 detonators. The TNT equivalence used was 1.32.

Each of the experiments utilised a grid of pressure measuring locations in the space behind the walls which extended from R = 0.15 m to 1.8 m, horizontally behind the wall, in intervals of 0.15 m and from h = 0.0 m to 0.9 m, vertically above the ground, also in intervals of 0.15 m. This provided a total of 84 measuring locations for each experimental configuration. Measurements were made using pressure transducers configured to measure side-on pressure-time records.

The database of the experimental results included a total of 517 individual cases. It should be noted that not all experimental records were included in the database for the development of the ANN models. The experimental database cases were visually inspected and some of the records that exhibited contradicting trends to the majority of the records in their group were removed from the database. The database used for the development of the ANN models comprises a total of 285 measurements.

In the majority of the applications of ANNs in various areas of engineering, the data are divided into sets needed to develop ANN models (e.g. training, testing, and validation) on an arbitrary basis without giving adequate attention to the statistical consistency of the data subsets. The cross-validation technique (Shahin et al. 2004) in which the training set is used to adjust the connection weights, the testing set is used to check the performance of the network and to decide when to stop training, and the validation set is used to evaluate the performance of the model is considered to be the most effective method to avoid problems with overfitting and generalisation of the network. Consequently, the statistical properties of the data subsets used in the model development need to be similar to ensure that each subset represents the same statistical population. If this is not the case, it may be difficult to justify the validity of ANN models (Shahin et al. 2004).

In this study, the available data were divided in a way that ensured that the statistical properties of the data subsets were close to each other and thus represent the same statistical population. The statistical parameters used include the mean, standard deviation, minimum, maximum, and range. As part of this approach, the 285 individual records were divided into three statistically consistent subsets. In total, 85% of the data

Recent advances in security technology

were used for calibration and 15% of the data are used for validation. The calibration set was further divided into 70% for training and 30% for testing.

Neural network implementation

Multi-layer perceptron networks trained by means of the back-propagating algorithm are used for the development of ANN models. The computer program Neuroshell2 (2000), Release 4.0, was used in the present study to train the network. The program implements several types of neural networks and architectures, including the multi- layer backpropagation nets, the genetic adaptive general regression neural networks (GRNN), the polynomial nets (GMDH), the backpropagation nets with jump connections and some others.

The optimum network architecture was obtained by evaluating the performance of ANNs with one, two, and three hidden layers and variable numbers of hidden layer nodes. In order to determine the optimal control parameters for the backpropagation algorithm, the networks were trained with different combinations of momentum and learning parameters.

BWAn – an engineering tool for predicting the effectiveness of blast wall barriers

The blast wall model developed in this paper is based on the artificial neural network methodology. It can be utilised for rapid prediction of the benefit of having a blast for a specified scenario. The computer program, BWAn, has been developed to allow the user to determine the pressure and impulse adjustment factors that can be used to modify the free-field pressures and impulses in order to predict the reduced level of pressures and impulses behind a blast wall. The program presents the results as contours of the adjustment factors for pressure and impulse. The user enters the blast threat parameters (standoff distance and charge weight) and the design height of a blast wall (see Fig.3). The program uses this information to generate adjustment factors for free-field positive pressure and positive impulse in the area behind the blast wall barrier. These reduced blast loads can be used in performing vulnerability and security planning analyses.

The purpose of BWAn computer program is to assist decision makers in determining the need for and the location of blast wall barriers. It is intended for architects, urban designers, structural engineers, and security consultants that are planning perimeter blast barriers for new and existing facilities where the risk to that facility justifies the expense of constructing the hardened reinforced concrete blast wall barrier.

Recent advances in security technology

Recent advances in security technology Figure 3. Screenshot of computer program BWAn Conclusion The main objective

Figure 3. Screenshot of computer program BWAn

Conclusion

The main objective of this study was to evaluate a new approach of using artificial neural network for predicting the blast environment behind a vertical blast wall barrier. A database of overpressures and impulses in the region behind the wall was established from the measurements of the blast environment during a series of small-scale blast wall experiments. This database was used to train and test the neural network to serve as a prediction tool for various blast wall scenarios. The main advantage of this approach is its ability to predict the effectiveness of a blast wall configuration in a matter of minutes whereas the numerical CFD simulation of the problem could take hours or days to complete. Analysis of the artificial network’s performance shows the feasibility of using ANN models for predicting non-ideal airblast loading in situations where the blast environment is modified by the presence of such rigid obstacles as protective walls and

Recent advances in security technology

multiple buildings. It has been demonstrated that the effectiveness of a blast wall can be quantified using the so-called ‘benefit’ contour plots of overpressure and impulse adjustment factors that are generated by the trained neural network model. The developed blast wall neural network model has also been deployed as a standalone application that could be used as a fast-running engineering-level tool within an expert system capable of predicting information about the likely injury and damage levels in blast environments where shielding by barrier walls is present.

References

Bogosian, D. & Shi, Y. 2002. Assessment of and prediction methodology for blast barrier effectiveness, Karagozian & Case, Report TR-01-34. Century Dynamics, 2005. AUTODYN-2D & 3D Version 6.0 User Documentation. Crepeau, J.E. 1999. SHAMRC, Second-order Hydrodynamic Automatic Mesh Refinement Code, Volume 1: Methodology, Applied Research Associates, Albuquerque, New Mexico. DAHSCWEMAN 1998. “Technical Manual – Design and Analysis of Hardened Structures to Conventional Weapons Effects; PSADS (Protective Structures Automated Design System), Version 1.0”, US Army Corps of Engineers, Washington, DC. Dove R. et al. 1989. Perimeter walls for blast reduction. Proceedings of the ASCE Specialty Conference on Structures for Enhanced Safety and Physical Security: 142-153. GSA, Office of the Chief Architect. ATBlast 2.2, General Services Administration, <http://www.oca.gsa.gov> Humphreys E. & Piepenburg D. 2001. Assessing effectiveness of blast walls: final report for the period 1 January 2000 to 31 December 2000, SAIC-01/1003, Science International Corporation. Mayor, R.P. & Flanders R. 1990. Technical manual, simplified computer model of air blast effects on building walls. US Department of Transportation, Research and Special Programs Administration, Transportation Systems Center, Cambridge. Mays, G.C. & Smith, P.D. 1995. Blast effects on buildings: Thomas Telford. NeuroShell2, Release 4.0, 2000. User’s manual. Ward Systems Group Inc., Frederick, MD. Remennikov, A.M. 2003. A review of methods for predicting bomb blast effects on buildings. J Battlefield Technology; 6(3): 5-10. Remennikov, A.M. & Mendis P. 2006. Prediction of airblast loads in complex environments using artificial neural networks. Proceedings of the 9 th International Conference on Structures Under Shock and Impact (SUSI 2006), 3-5 July, The New Forest, UK, pp 269-278. Rice, D.L. et al. 2000. Experimental and numerical investigation of shock diffraction about blast walls. Proceedings of the 16 th International Symposium on the Military Aspects of Blast and Shock (MABS- 16), Keble College, Oxford, U.K. Rose, T.A. et al. 1995. The effectiveness of walls designed for the protection of structures against airblast from high explosives. Proc. Instn Civ. Engrs Structs & Bldgs 110: 78-85. Rose, T.A. 2001. An approach to the evaluation of blast loads on finite and semi-finite structures, PhD Thesis, Cranfield University, Royal Military College of Science. Rose, T.A. 2003. Air3D User's Guide. 7.0: RMCS, Cranfield University, UK. TM5-583-3, 1986. “Security Engineering Manual”, Department of the Army, Washington, D.C. TM5-1300, 1990. “Design of structures to resist the effects of accidental explosions”, U.S. Department of Army Technical Manual, Washington, D.C. Shahin, M.A. et al. 2004. Data division for developing neural networks applied to geotechnical engineering. J Computing in Civil Engineering, ASCE; April: 105-114.

Recent advances in security technology

8

Blast Reliability Curves and Pressure-Impulse Curves:

Complementary Information for Decision Makers.

M.D. Netherton & M.G. Stewart

University of Newcastle, Australia.

Abstract

Pressure-Impulse (P-I) curves detail which combinations of explosive blast load and impulse will cause a specified level of damage. However, the factors used in the formulation of these curves are deterministic in nature and do not take into account the uncertainty and variability of explosive loads or structural response. A Blast Reliability Curve (BRC), which considers uncertainty and variability of blast load and structural response, shows probabilities of damage associated with various blast loads. Using a related set of load-response parameters applicable to typical glazing systems, this paper shows the development of P-I curves and Blast Reliability Curves. The information derived from each type of curve is compared, with assessments made on the utility of each and their complementary nature for decision makers.

Biographies

Michael Netherton is a Post-Graduate research student in the Department of Civil, Surveying and Environmental Engineering at The University of Newcastle, Australia. He received his MSc in Weapons Effects on Structures in 2000 from RMCS and when a Squadron Leader with the RAAF was the lead Australian on the Coalition's Weapons Effectiveness Assessment Team in post-war Iraq 2003.

Recent advances in security technology

Mark Stewart is a Professor in the Department of Civil, Surveying and Environmental Engineering at The University of Newcastle, Australia. He received his PhD in 1988 from The University of Newcastle. His research interests include structural and serviceability reliability, stochastic deterioration modelling, security risk assessment, uncertainty modelling, decision analysis and quality management issues for design and construction processes.

1.

Introduction

Pressure-Impulse (P-I) curves (or iso-damage curves) describe which combinations of explosive blast pressure and impulse may cause damage (e.g., Baker, et al. 1983, Smith & Hetherington 1994 and Krauthammer 1998). P-I curves provide information on whether a damage threshold is reached, and so provides for discrete levels of damage outcomes ranging from fail/no-fail to more refined levels such as no damage, minor, moderate, etc. In some cases P-I curves are derived from empirical (field trial) data which may be difficult to correlate to other circumstances. A P-I curve is a deterministic tool, in that, the variables used have given or set values and the outcome is in terms of discrete levels of damage; for example, fail/no-fail. In this case, the outcome is either: (i) 100% probability of damage or (ii) 100% probability of no damage. There is no indication of likelihood or extent of damage other than the extremes of everything damaged or nothing damaged. The P-I curves are also likely to be conservative (over-predict actual damage), which is a reflection of most decision- maker's preference to risk averseness. Notwithstanding, P-I curves remain very useful in allowing decision makers to quickly assess expected damage for given threats.

For any threat scenario there will be uncertainty and variability in blast load and structural response. A Probabilistic Risk Assessment (PRA) can propagate these uncertainties through structural response models to provide estimates of likelihood and extent of damage. Stewart and Netherton (2005a), Stewart, et al. (2006) & Stewart (2006) detail how such a probabilistic analysis can be used to develop Blast Reliability Curves (BRCs) which shows structural – or elemental – probabilities of failure associated with various blast loads on typical glazing systems. This provides decision makers with additional information which is not available from P-I curves or qualitative risk assessments based on risk-ratings, risk-scores or other non-probabilistic measures of risk.

In the present paper, the principle of P-I curve development will be shown – appropriate to a pane of monolithic glass (summarised from Smith & Hetherington 1994). Then, with respect to a particular pane of glass, the process for deriving Blast Reliability Curves (BRCs) will be shown (summarised from Stewart & Netherton 2005a & Stewart 2006). Finally, the information from the P-I curve and the BRCs, applicable to the common pane of glass, will be transposed to the facade of a 20 story building to show the utility of each; with an assessment made of their complementary nature for decision makers.

Recent advances in security technology

  • 2. SDOF Model of Structural Response

The determination of a structure's dynamic response to explosive blast loads via the Single Degree of Freedom (SDOF) model is well described (Biggs 1964, Baker, et al. 1983, TM5-1300 1990, Bangash 1993 and Smith & Hetherington 1994). A reason for the popularity of the SDOF model is given thus: "Because the SDOF model is simple and efficient and realistically represents the structural behaviour based on an anticipated mode of response, it is a useful tool to predict the overall response of a structure, which determines the overall damage level of a structural system or structural element." (Li & Meng 2002).

Following the work of Baker, et al. (1983), Smith & Hetherington (1994) consider a structure which has been idealised as a SDOF elastic structure and subjected to a blast load (idealised as a triangular pulse) delivering peak force F and positive phase duration of t d . See Figure 1.

Recent advances in security technology 2. SDOF Model of Structural Response The determination of a structure's

Figure 1. Single degree of freedom elastic structure subject to idealised blast pulse. where M is the structure's mass, c is the damping coefficient and K is the structure's elastic response.

In the consideration of a structure's response to a given blast load, it is the calculation of the displacement of a structure – or its elements – that is most often desired; where a maximum displacement (xmax) may be specified as a limit-state or failure criteria. See Stewart & Netherton (2005b) for further details on the derivation of structural displacement via the SDOF model.

  • 3. Pressure-Impulse Curves

From considerations of different positive phase durations of the idealised blast load and the natural period of the structure – as well as equating Strain Energy with Work Done and Kinetic Energy – Smith & Hetherington (1994) produce a graphical representation of quasi-static, impulsive and dynamic response possibilities. They then convert this representation to a "pressure-impulse diagram which allows the load-impulse combination that will cause a specified level of damage to be assessed very readily". Different types of structures loaded by different blast load profiles result in different P-

Recent advances in security technology

I diagrams. For a pane of monolithic glass modelled as a SDOF (see Figure 1), the appropriate P-I diagram is depicted in Figure 2.

Recent advances in security technology I diagrams. For a pane of monolithic gl ass modelled as

Figure 2. Non-dimensionalised pressure-impulse diagram for SDOF elastic system. (after Baker, et al. 1983)

Once a particular structure's P-I curve has been calculated and drawn, to then plot a point relative to the hyperbolic curve, the user has to enter values of blast pressure (F), impulse (I), structural stiffness (K), mass (M) and maximum displacement (x max ). Any point plotted to the left and below the curve describes a displacement less than the limit-state; whereas, any point plotted above and to the right of the curve describes a displacement greater than the maximum allowed and thus a failure is deemed to have occurred. As can be seen, if a particular group of variables are selected, then it is possible to plot a point that describes a displacement that exactly equals the maximum displacement allowed; any such point lies exactly on the P-I curve, the implications of which will be discussed later in an illustrative example.

  • 4. Blast Reliability Curves

A P-I curve assumes that all input parameters associated with the loading, the structure and its response are known with certainty; however, in reality, there may well be considerable uncertainty regarding these values. When consideration is given to this issue, the problem is no longer the simple determination of limit state exceedance; rather, it becomes a structural reliability matter with the calculation and prediction of the probability of limit state violation (Melchers 1999). These matters of parameter and model uncertainty are the key to the development of Blast Reliability Curves.

Threat scenarios are subject to considerable uncertainty, as are system response, effectiveness of protective measures and expected damage. Probability theory can be

Recent advances in security technology

used to represent such uncertainty and it is such uncertainty of outcomes that leads to ‘risk’; such as damage, fatality or economic risk. The treatment of probabilistic measures of uncertainty in a risk analysis produces a Probabilistic Risk Assessment (PRA) (e.g., Stewart & Melchers 1997).

Risk is the product of probability of occurrence (or failure) and consequences (p f xC F ). This definition is consistent with the interpretation of risk used in most engineering systems, Risk Management standard AS/NZS 4360 (2004) and the insurance industry (expected losses). Risk may be given in terms of dollars, number of human fatalities, etc. such as expected damages per year or expected fatalities per year. A general description of probabilistic risk analysis and assessments for civil engineering and other engineering systems is provided elsewhere (Stewart & Melchers 1997, Faber & Stewart 2003). Clearly, if there is 100% certainty of information and outcomes then the probability of failure is either 0.0 or 1.0. However, such certainty in outcomes is rare.

There is uncertainty about:

  • a) Input data – expert predictions, material strengths, detonation charge, stand- off, etc.

  • b) Accuracy of predictive models – computer and consequence models, threat scenarios, etc.

  • c) Inherent variabilities – weather, individual exposure to hazard, time of blast,

etc. This all leads to uncertainties associated with the magnitude of a blast, its likelihood and the consequences. A PRA propagates these uncertainties through the computations to reveal an estimate of risk. For a simple component of strength R subject to loading effect S the probability of failure is:

p

f

= Pr( R S ) = Pr( R S 0) = Pr [G ( R, S ) 0]

(1)

where G() is the limit state function and G()=0 defines the boundary between the ‘unsafe’ and ‘safe’ domains. Usually, predictive models of system loading and response are incorporated into the limit state function(s). In general, the resistance and load effect are dependent on a vector of variables X and the limit state function represented as G(X). One of the ways that a reliability analysis of blast damage to built infrastructure can be represented is a probability of failure conditional on occurrence of a specific threat scenario; thus:

p

f

Θ = Pr

ij

[G

( X ) < 0 |

S

=

s ] [S

Pr

=

s ]

(2)

where Θ ij is the threat scenario for a specific explosive weight. Further, Pr(S=s) represents the probability distribution of blast loading for a specific threat scenario (ie. known explosive weight and stand-off distance) considering inherent, model error and parameter uncertainties.

Recent advances in security technology

See Stewart & Netherton (2005b) for complete details of the derivation of Eqns. (1) and (2).

Through the use of Eqn (2), a ‘Blast Reliability Curve’ (BRC) can be produced, and is a convenient way to summarise conditional probabilities of failure where one variable is fixed and the other given a range of values. For example, a blast reliability curve can be a plot of probability of failure vs. stand-off distance, for a specific explosive weight; an illustrative example will be given in the next section.

  • 5. Blast Reliability Curve for a Single Pane of Glass

Consider a pane of monolithic annealed glass with the following parameters:

  • (a) Height = 2000mm, width = 2000mm, thickness = 10mm;

  • (b) Density = 2500 kg/m 3 ;

  • (c) Young's modulus = 6.9x10 10 N/m 2 ; and

  • (d) Maximum tensile stress permitted = 84.8 MPa.

The pane of glass can be considered as a plate in the x-y plane that (when simply- supported on all four sides and subject to a blast load) experiences maximum moments in the pane's centre; which describe peak stresses in the x and y directions, which in turn, describe a maximum deflection at the centre of the pane. (Krauthammer & Altenburg 2000). Thus, using the deflection that causes a maximum tensile stress of 84.8Mpa as a limit state, and considering the glass as a Single Degree of Freedom elastic structure simply supported within a window frame and then subject to the idealised blast pulse from Figure 1 – from say 50kg TNT – and using Eqn (2), a specific BRC can be derived (Netherton & Stewart 2006), see Figure 3.

Recent advances in security technology

Recent advances in security technology Figure 3. Blast Reliability Curve for a 2m x 2m annealed

Figure 3. Blast Reliability Curve for a 2m x 2m annealed glass pane (10mm thick) subject to

50kgTNT.

  • 6. Illustrative Example: 20 Story Building

To demonstrate the utility and differences between P-I curves and BRCs, consider a typical 20 storey commercial building 76 m high and 35 m wide (following Netherton & Stewart 2006). The facade comprises 2 m × 2 m windows (simply supported on all four sides), 17 windows per floor, each window being separated horizontally by 50 mm and vertically by 1800 mm. According to the Australian glazing design and wind loading codes (AS/NZS 1288 (2006), AS/NZS 1170.2 (2002)), one design solution for these windows is 10 mm annealed glass.

Within this example, the unique obliquity of each window relative to an explosive event is considered in any calculation of blast pressures and impulses impinging on the glass panes. Further, the 20 story structure is considered as part of a continuous down- town environment, thus, it is assumed to be part of an infinitely wide facade; as such, the effects of clearing are not considered.

Within Section 3, it was noted that it is possible to plot a point that lies exactly on the P-I curve. Let us presume that an amount of explosive charge – say 50kg TNT – is located on the ground and at a distance perpendicular to the front facade of the 20 story structure. Then, if we presume that the point lying exactly on the P-I curve commensurately describes a point-in-space located somewhere along the vertical centre-line of the facade, thus, by the spherical nature of the blast’s emanating shock

Recent advances in security technology

wave, there must be an arc of similarly exact P-I curve points-in-space located across the facade of the building. See Figure 4(a). If the 50kg of TNT is located closer to the structure, then the arc of points must move up the face of the facade, and if the explosive is located further away, the arc necessarily moves lower. Regardless of the position of the arc relative to the distance of the explosive from the facade, according to the fail/no-fail methodology of P-I curves, any window in the region below the arc- line will be predicted to fail, and any in the region above will be predicted to stay intact.

It was shown in Figure 3, for a given window, there are different probabilities of failure conditional on occurrence of a specific threat scenario. Staying with a weight of 50kg TNT, and fixing the location of the explosive at a defined point on the ground and at a distance perpendicular to the front facade of the 20 story structure – say 50m – it has been shown by Netherton & Stewart (2006) that Blast Reliability Curves can be developed for each window in the building’s facade. As stated earlier, because the angle of incidence and distance to source of detonation varies for each window, this means that the peak reflected pressure and impulse for each window also changes. The BRCs are then used to produce contours of glazing damage risks across the facade of the building (see Figure 4.b).

With regard to the BRC risk contours, it means that those windows at the top of the structure have approximately 15% probability of damage, whereas those closer to the bottom floor – and thus closer to the explosive charge – have an increased probability of damage. In this example, those windows at the centre of the bottom floor have at least a 95% chance of failure.

Recent advances in security technology

Recent advances in security technology (a) Arc of similar P-I curve points and Fail/No-fail (b) BRC

(a) Arc of similar P-I curve points and Fail/No-fail

Recent advances in security technology (a) Arc of similar P-I curve points and Fail/No-fail (b) BRC

(b) BRC risk contours

regions Figure 4. Comparison of Glazing Damage to Building Facade, for W=50 kg TNT.

  • 7. Comparison of P-I curve and BRCs

Clearly, there is significant difference between the information presented by the P-I curve's fail/no-fail information and the Blast Reliability Curve risk contours. The P-I curve provides an arc-line across the building which presumes that all glass below the line fails; whilst all that above does not. In certain circumstances this level of information (or lack of it) may be appropriate when a quick idea of "safe" distances is required; however, experience from real blasts against similar structures (i.e. those buildings adjacent to the Australian Embassy bombing in Indonesia in 2004) shows that there is no definitive fail/no-fail line. Rather, there is a wide zone where the breakages phase from complete breakage regions to no breakage regions, i.e., the extent of damage is highly spatially variable. This real world experience is modelled more closely by the BRC risk contours, where the probability of window of failure (conditional on a specific threat scenario) decreases as the slant range increases up the face of the structure.

P-I curves are less computationally demanding for the user, in that, once a specific curve has been derived for a given structure (or its element) then the work of inputting appropriate parameters (leading to an answer of fail/no-fail) is relatively quick as compared to using a BRC. A pre-calculated P-I curve also accommodates a wide variety of pressure and thus describes the entire range of explosive weights and ranges.

Recent advances in security technology

As compared to BRCs, which must be computed for each element and at least one of the blast variables.

  • 8. Complementary information for Decision Makers

The illustrative example in the present paper only deals with a specific piece of glazing as used across the facade of a typical building; however, it is clear that the differences between the information provided by P-I curves and BRCs will similarly translate to all manner of explosive blast loads against a range of other structures and/or structural elements.

P-I curves, once created for a structure (or its elements) provide a very quick method for determining the boundary between fail and no-fail regions. Decision makers may feel that this information is sufficient for their needs. BRCs, however, require more effort to produce the number of curves which would adequately cover the range of possibilities available from within a single P-I curve. Notwithstanding the extra effort required for creation, BRCs can also be produced in advance for prompt risk evaluation by decision makers, examples of their use include:

  • (a) Post-blast forensics. Police and other on-scene investigators may take measurements of broken windows, compare the zone of damage against probabilities of failure via BRCs, and thus back-calculate probable blast weights.

  • (b) Collateral Damage Estimation. The risk of creating a given level of damage (against glass, structures and humans) is not given by a P-I curve. Consequently, decision makers can be well served by the additional information regarding conditional risks (and expected losses) that are available from BRCs.

  • (c) Risk mitigation advice. A P-I curve does not quantify the reduction in risk for different blast mitigation treatments. These matters, particularly including Life Cycle Costing, are a feature of BRCs and are well described by Stewart, et al. (2006) and Stewart (2006).

  • 9. Conclusion

The present paper described how Pressure-Impulse curves and Blast Reliability Curves can be created; with information from each curve assessed in terms of panes of window glass as used on a 20 story commercial building. It was shown that whilst Pressure- Impulse Curves provide a level of useful information, a Blast Reliability Curve can provide additional and complementary information that may be of further benefit to decision makers.

Acknowledgements

Recent advances in security technology

The assistance provided by Mr Mark Arkinstall during discussions about P-I curve generation is appreciated and also Dr Xiaoli Deng for her assistance in preparing Figure 4(b). The support of the Australian Research Council under grant DP0556913 is gratefully acknowledged.

References AS/NZS 4360, 2004. Risk Management. Standards Australia. Strathfield, NSW.

AS/NZS 1288, 2006. Glass in Buildings - Selection and Installation. Standards Australia.

Strathfield, NSW.

AS/NZS 1170.2, 2002. Structural Design Actions – Part 2: Wind Actions. Standards

Australia. Strathfield, NSW.

Baker, W. E., Cox,

P. A., Westine, P.

S.,

Kulasz,

J.

J.,

Hazards and Evaluation. Elsevier, Amsterdam.

& Strelow, R.

A. 1983. Explosion

Bangash, M.Y.H. 1993. Impact and Explosion – Analysis and Design. Blackwell Scientific Publication, Oxford. Biggs, J.M. 1964. Introduction to Structural Dynamics. McGraw-Hill, New York.

Faber, M.H. & Stewart, M.G. 2003. Risk Assessment for Civil Engineering Facilities: Critical Overview and Discussion. Reliability Engineering and System Safety. 80(2): 173-184.

Krauthammer, T. 1988. Blast mitigation technologies; developments and numerical considerations for behavior assessment and design. Proceedings of the International Conference on Structures Under Shock and Impact. SUSI, Computational Mechanics Inc., 3-12. Krauthammer, T. & Altenburg, A. 2000. Negative Phase Blast Effects on Glass Panels. International Journal of Impact Engineering. 24(1): 1-17. Li, Q.M. & Meng, H. 2002. Pressure-Impulse Diagram for Blast Loads Based on Dimensional Analysis and Single-Degree-of-Freedom Model. Journal of Engineering Mechanics. Vol.128, No 1, January 1, 87-92.

Melchers, R.E. 1999. Structural Reliability Analysis and Prediction. 2nd ED, John Wiley and Sons, New York. Netherton, M.D. & Stewart, M.G. 2006. Security Risks and Structural Reliability of Window Glazing Subject To Explosive Blast Loading. 3 rd International ASRANet Colloquium. Glasgow, UK. Stewart, M.G. & Melchers, R.E. 1997. Probabilistic Risk Assessment of Engineering Systems. Chapman & Hall, London. Stewart, M. G. & Netherton, M. D. 2005a. Blast Reliability Curves and Uncertainty Modelling For Glazing Subject To Explosive Blast Loading. 6th International Conference on Shock & Impact Loads on Structures. Hao, H., Lok, T.S., & Lu, G.X (Eds). C I Premier Conferences, Singapore, 523-530 Stewart, M. G. & Netherton, M. D. 2005b. Security Risks, Uncertainty Modelling and Probabilistic Risk Assessment of Glazing Subject to Explosive Blast Loading. Centre for Infrastructure Performance and Reliability. The University of Newcastle, Australia. Research Report No 255.10.2005. Stewart, M. G., Netherton, M. D. & Rosowsky, D. V. 2006. Terrorism Risks and Blast Damage to Built infrastructure. Natural Hazards Review. v 7, n 3, 114-122. Stewart, M. G. 2006. Risk Assessment and Optimisation of Blast Mitigation Strategies for Design and Strengthening Built Infrastructure. Keynote Paper. 1st International Conference on Analysis and Design of Structures against Explosive and Impact Loads. Tianjin, China.

Recent advances in security technology

Smith, P.D. & Hetherington, J.G. 1994. Blast and Ballistic Loading of Structures. Butterworth- Heinemann, Oxford, UK.

TM5-1300, 1990. Design of Structures to Resist the Effects of Accidental Explosions, US Department of the Army Technical Manual TM5-1300, USA.

Recent advances in security technology

9

Response of Plates to Blast Loading in Inelastic Range

Gregory Szuladzinski

Analytical Service Pty Ltd, Killara/ Sydney, Australia

Abstract

Several classical support conditions are investigated for plates with uniformly distributed loading, in the form of a pressure pulse. Both circular and rectangular plates are described. Bending and shear modes are considered. First, a simple calculation is presented, using algebraic equations of momentum and energy conservation to derive approximate deflection results. Then, exact relationships, derived from work by others are quoted and discussed. Those relations were developed using a manageable, rigid- plastic material model, which results in underestimating true deflections. When numerical solutions are obtained by employing elastic, perfectly-plastic material model, it is found that the simpler solutions proposed here provide more accurate results of final deflections. A detailed numerical study is conducted using a circular plate example.

Biography

He received his Masters Degree in Mechanical Engineering from Warsaw University of

Technology in 1965 and Doctoral Degree in Structural Mechanics from University of Southern California in 1973. From 1966 to 1980 he worked in the United States and then in Australia in a range of industries and applications. He has a number of publications to his credit in stress analysis, dynamics and plasticity. His book entitled

“Dynamics of Structures and Machinery.

Problems and Solutions” was published

worldwide by John Wiley Interscience in 1982. He is a Fellow of the Institute of

Engineers Australia and a member of the ASME.

Recent advances in security technology

Introduction

Structural loading resulting from explosive action will be briefly discussed with reference to Fig. 0. The explosion itself takes place at t = 0. After the arrival time t a an observer at some distance from the explosion source can see pressure rising, almost instantly, to the level of p s , usually called side-on pressure. Then, over the interval t s the overpressure drops to zero, or pressure drops to the atmospheric level p a . This ends the positive phase of the pressure history. The decrease of pressure may continue in time so that the negative phase is entered with the peak suction Δp reaching a fraction of the atmospheric pressure level.

In estimating the structural effects the attention is usually focused on the positive phase. The first reason for it is that at some range of distance from a source the negative phase may not develop. The second is that the Δp is usually much smaller than p s , although this argument is offset by a long duration of the negative phase. The side-on impulse i s is the hatched area in Fig. 0a.

The pressure history in Fig. 0a is a free-field diagram,which means there is no significant interference between the moving blast wave and the objects in its way. For structural loading, however, this is only a half of the story. When a plate, for example, is impacted by the wave, the reflected pressure is what matters, rather than the free- field pressure in (a). Furthermore, the duration of that pressure is just as important. This is another way of saying that what really matters is the reflected impulse i r , or the area under reflected pressure, when plotted against time.

The blast wave information, such as pressures, durations and impulses are available in a number of documents, one of them being TM5 (1992). Those parameters serve to create simplified pressure diagrams, typically in the form of a triangular decreasing pulse, such as one in Fig.0b. (A more involved diagram may be needed, if the natural period of the impacted structure is long enough to be comparable to t s .) The rectangular pulse (with the same shaded area) in (c) used instead results in only a small decrease in accuracy, especially if pulse duration is much smaller than the natural period. (This response to this pulse is often a convenient reference and for this reason it is used exclusively in this presentation.)

One should also mention that the reflected pressure p m is typically several times larger than the side-on pressure p s and the same relation applies to impulses, as i r > i s . When a plate is subjected to loads of large magnitude, but short duration, the anticipated failure is often related to the peak deflection or, sometimes, to the maximum slope.

In order to preserve simplicity of a theoretical formulation it is necessary to replace an actual stress-strain curve of material (Fig.1) with a simplified shape. The most popular approximations, when large strains are involved, are the elastic - perfectly plastic (EPP) and rigid-plastic (RP). The latter one represents a material, which is assumed to be undeformable up to a certain stress level σ o , often called the flow stress. When the

Recent advances in security technology

level is reached, further elongation takes place with no increase of resistance. Based on elastic strain magnitude alone, simplifying the EPP to RP model seems almost natural. Yet, this simplification has consequences not always easy to predict in advance. While RP makes calculations easier, it is definitely less realistic than the EPP model.

The detailed solutions for selected edge restraint cases were developed, or presented on the basis of work of others, by Jones [1997]. Those were obtained by integrating the equations of motion. A simpler, energy-momentum approach presented below offers a substantial mathematical simplification. The main objective of this work is to establish how realistic deflection predictions one can obtain with the existing formulas, as compared to a numerical approach.

Limit Values of Shear Force and Bending Moment

The bending moment capacity of a plate is related to the capacity of a rectangular beam of a unit width, such a fictitious beam being a part of a cross-section of a plate. As bending moment grows, a state is attained when the outer fiber stress reaches σ 0 , the onset of yield. The associated bending moment is M y . With a further load growth a fully developed plastic stress distribution is reached and the moment attains its maximum capacity of M 0 . According to the basic theory:

M y = I σ 0 /c

and

M 0 = Z σ 0

(1a,b)

where I=h 3 /12, c= h/2 and Z are section properties. (Note that the first one is the basic bending formula.) For plates, where the sections are rectangular, with a unit width and thickness h, constant Z equals h 2 /4 the above simplifies to

M y

=

  • 1 h s

2

and

M 0 =

1

2

h s

0

  • 6 4

  • 0 (2a,b)

The yield shear stress, or flow stress in shear, σ 0s = σ 0 /

Recent advances in security technology level is reached, further elongation takes pl ace with no increase

3 , is consistent with energy-

of-distortion theory. The corresponding onset of yield in shear and the shear capacity are, respectively

Q y = (h/k) σ 0s

and

Q o = h ef σ 0s

(3a,b)

where h ef is the section area of the unit-width, effective for the limiting condition while h is the total area of this cross-section. The ratio of peak elastic shear stress to the average shear stress, designated by here by k can be found, for example, in Roark [1975]. For a rectangle k = 3/2 and h ef h.

Recent advances in security technology

Elastic, Static Response To Uniform Loading

There are four plates under consideration, as shown in Fig. 2: circular, square, rectangular (2/1 side ratio) and infinitely long. Each of them can have the edges supported or clamped. When a uniform load w (N/m 2 ) is applied to the surface, the peak bending moments and center deflections can be expressed as:

M = b wa

2

or

M = b wR

2

(4a,b)

where a is a shorter side of a rectangular plate and R is the radius of a circular plate. The deflection at center is

wa

44

wR

or

g

Eh

3

uu

g

==

D

D

1 2 (1

-

n

2

)

with

D

=

(5a,b,c)

The coefficients β and γ are given in Table I. With shear forces the situation is simpler in that the distribution is regarded as the same for supported as it is for clamped edges (just as for beams). For a circular plate the equilibrium condition tells us that Q = Rw / 2 . For a square plate, according to Roark [4], Q = 0.42wa, for the 2:1 side

ratio Q = 0.503wa 0.5wa and for the infinite side ratio, Q = 0.5wa. (Note that if the distribution were uniform in case of the square plate, the coefficient would be 0.25 rather than 0.42.) The maximum shear load (per unit length) can therefore be written as

Q = j wa

or

Q = j wR

where the α coefficients are in Table I.

CIRC/SS

CIRC/CL

Square/SS

Square/CL

2:1/SS

2:1/CL

INFIN/SS

INFIN/CL

β

3

+ n

16

1/8

0.0479

0.0513

0.1017

0.0829

1/8

1/12

φ

1/2

1/2

0.42

0.42

1/2

1/2

1/2

1/2

γ

5 + n

64(1

+ n

)

1/64

1/245.9

1/791.3

1/98.38

1/394.2

1/76.8

1/384

ξ

6

12

24

48

(42.84)

12 (14)

24

8

16

(24.14)

J

1/6

1/12

1/24

1/48

1/15

1/30

3/32

3/64

(6a,b)

Table I Deflection and strength coefficients

Static Onset Of Yielding

Setting the value of yield moment from (2a) in (4a,b), the following load, associated with the onset of yield in bending is obtained:

w

y

=

s

0

2 æö h ç ÷ ç ÷ ç ÷ èø a
2
æö
h
ç
÷
ç
÷
ç
÷
èø
a

s

0

æ

ç

b è

ç

h

6

b

6

ç

R

and

w

y

=

2

ö

÷

÷

÷

ø

Recent advances in security technology

(7a,b)

Similar procedure applies for shear, where combining Eqs. 3a with 6a,b and setting h ef = h and k = 3/2 one gets:

w

y s

=

2

s

æö h 0 s ç ÷ ç ÷ ç ÷ èø a
æö
h
0 s
ç
÷
ç
÷
ç
÷
èø
a

2

s

0 s

h

æ

ç

è

ç

3

j

3

j

ç

R

and

w

ys

=

ö

÷

÷

÷

ø

(8a,b)

for rectangular and circular plates, respectively.

Static Collapse Load

The reasoning applied to derive the collapse load is purely static, keeping in mind that there is no deformation until the load attains that limiting value. When a collapse begins, yielding is postulated to take place along dashed lines, as in Fig.2. At this instant the external pressure applied to a plate fragment is balanced by the support reactions as well as by fully developed internal forces. (This is for simply-supported plates. When the edges are clamped, the yield lines also appear along the edges.) The simplest case is that of a circular plate, as in Fig.3a,b, where a radial segment, with bending moments along radial sections is depicted. For a small central angle α the wedge is approximated by a triangle. The resultant force is αR 2 w b /2 and this force is located at R/3 from the edge. The moment of loading about the edge tangent is balanced by the internal moment, also projected onto that tangent:

1

R

æ

÷

÷=

÷

ø

ö

6

M

0

M

0

a

R

2

ç

3

R

2

a

Rw

b

ç

è

ç

2

or

w b
w
b

=

(9a,b)

For a clamped edge there is the same internal moment along the circumference, resisting pressure as well. This doubles the collapse load w b . For the remaining shapes the procedure is similar. For the square shape, which fails along diagonals, one-quarter of a plate is shown in Fig.3c along with the internal bending moment. The results for all plates can be presented as

w b = ξ M 0 /a 2 w b = ξ M 0 /R 2

(10a,b)

for rectangular and circular plates, respectively. The results obtained from such simple schemes should be related to the accurate results provided by Jones [1]. The coefficients ξ are listed in Table II and, in case of discrepancy, the results of the accurate method are given in brackets. With regard to the infinitely long plate, the procedure is to consider a unit-width beam spanning the supports. The limiting loads are the same as for the corresponding beams.

Recent advances in security technology

The determination of collapse loads in shear mode is made easier by assuming the same distribution of internal shears along the edge in both elastic and inelastic ranges. Equating the applied shear load Q from (6) with shear capacity from (3b) results in the following loads being associated with shear collapse by sliding:

w

s

=

s

0

s

æö

h

ç

÷

÷

÷

èø