Académique Documents
Professionnel Documents
Culture Documents
Operations Management-I
Submission Date: nd 22 Dec 2012
Minor Project
Weightage: 10%
Acknowledgement
We would like to extend our heartfelt thanks to Dr. Manoj K. Srivastava for giving us the opportunity to work on a challenging topic such as this and guiding us throughout. Though We faced a lot of difficulties in narrowing down our scope to be able to cover under this project, we were given some helpful ideas which gave us a direction to work in. This project has increased our understanding of cloud computing and forecasting.
Abstract
Today, irrespective of industry, corporations must focus on speed, efficiency and customer value to be globally competitive. But most of all, the companies should focus on technological innovation to keep ahead of their competitors. Cloud Computing Techniques have recently taken the software world by storm and have been on top of every software makers wishlist. Cloud computing helps to reduce capital expenditures for an organization giving it the flexibility to feed inputs from any location to determine the best possible forecast.
This project aims to establish the benefits of cloud computing in supply chain management and understand different types of forecasting, What is a good forecast and evaluating whether the forecast model gives a good forecast. We have also developed a prototype model of Moving Average Method which forecasts the expected demand using the inputted data, calculating the performance of forecast model.
Literature Review
Cloud Computing
Cloud computing is the use of computing resources (hardware and software) that are delivered as a service over a network (typically the Internet). The name comes from the use of a cloudshaped symbol as an abstraction for the complex infrastructure it contains in system diagrams. Cloud computing entrusts remote services with a user's data, software and computation. In the business model using software as a service (SaaS), users are provided access to application software and databases. The cloud providers manage the infrastructure and platforms on which the applications run. SaaS is sometimes referred to as on-demand software and is usually priced on a pay-per-use basis. SaaS providers generally price applications using a subscription fee.
Experts claim that the SaaS allows a business the potential to reduce IT operational costs by outsourcing hardware and software maintenance and support to the cloud provider. This enables the business to reallocate IT operations costs away from hardware/software spending and personnel expenses, towards meeting other company goals. In addition, with applications hosted centrally, updates can be released without the need for users to install new software. One drawback of SaaS is that the users' data are stored on the cloud providers server. As a result, there could be unauthorized access to the data. End users access cloud-based applications through a web browser or a light-weight desktop or mobile app while the business software and user's data are stored on servers at a remote location. Proponents claim that cloud computing allows enterprises to get their applications up and running faster, with improved manageability and less maintenance, and enables IT to more rapidly adjust resources to meet fluctuating and unpredictable business demand.
Cloud computing relies on sharing of resources to achieve coherence and economies of scale similar to a utility (like the electricity grid) over a network. At the foundation of cloud computing is the broader concept of converged infrastructure and shared services.
Preventative Controls : These controls upgrade the strength of the system by managing the vulnerabilities. The preventative control will safeguard vulnerabilities of the system. If an attack were to occur, the preventative controls are in place to cover the attack and reduce the damage and violation to the system's security. Corrective Controls : Corrective controls are used to reduce the effect of an attack. Unlike the preventative controls, the corrective controls take action as an attack is occurring. Detective Controls : Detective controls are used to detect any attacks that may be occurring to the system. In the event of an attack, the detective control will signal the preventative or corrective controls to address the issue. Manufacturers need to make sure that cloud computing companies meet industry standards for security such as ISO 27001. 2. Legal Issues : Numerous regulations pertain to the storage and use of data, including Payment Card Industry Data Security Standard (PCI DSS), the Health Insurance Portability and Accountability Act (HIPAA), the Sarbanes-Oxley Act, among others. Many of these regulations require regular reporting and audit trails. Cloud providers must enable their customers to comply appropriately with these regulations. 3. Reliability : The cloud computing offering services have to make sure that they have redundant backups because for a 99% uptime The downtime is 3.65 days per year OR 7.20 hours in a month. As Bob Parker, group vice president of research for IDC Manufacturing Insights puts it : If Salesforce.com goes down for two hours, your salespeople will be inconvenienced, but it is not necessarily going to bring your company to its knees, said Bob Parker, group vice president of research for IDC Manufacturing Insights. But if you cant send invoices or process orders for half a day, youve got a different issue. 4. Customization :
Because users typically share the same instance of cloud computing software, the application may not be as customizable as on-premises software or traditional hosted ERP.
Cloud computing is focus of nearly all the top 20 SCM software suppliers. Some such as SAP are focussing on private cloud while others such as IBM on public cloud network, But nearly all are focussed on increasing share of revenues derived from cloud computing.
According to ARC Advisory Group study: Transportation Management System (TMS) derived nearly 25% of revenues from public or private cloud service Warehouse Management Systems (WMS) derived only 3% of revenue from cloud offerings This difference is mostly due to established Enterprise Resource Planning Vendors like Oracle and SAP integrating Warehouse Management Solutions in their products, exposing their APIs and providing integration hooks. But still for new businesses it is much cheaper to purchase SaaS WMS per warehouse at 500$ per month than invest Hundred thousand dollars and more in traditional WMS systems.
Forecasting
Forecasting is simply estimating future value of any variable. The variable can be demand for goods, supply for goods, prediction of share prices. General laws of forecasting: 1. Forecasting involves risk : 2. Forecasting provides a future value which may exceed the actual value or be less than the actual value. 3. Forecast are generally more accurate in near term than long term : The longer the duration of prediction, the more difficult it becomes to predict the demand because of increasing number of factors. 4. Forecast can never be substitute for calculated value : If a company can get actual data for expected demand through efficient IT use. It would not need forecasting. If forecasting does not give exact value, Why forecast? Forecasting is done because it reduces inventory costs as a reward for risks. Thus forecasting cost should be always less than the risks involved. So Managers are deeply focussed on Returns on Investments made in forecasting.
Types of forecasting
The two main types of forecasting are 1. Qualitative forecasting 2. Quantitative forecasting
Qualitative forecasting
Qualitative forecasting is subjective in nature, taking into account the views of consumers and experts. It is most appropriate when past data is not available. Example of Qualitative forecasting is Delphi Method, Market research.
Quantitative forecasting
Quantitative forecasting is an objective method based on availability of numerical past data to forecast future value. Some of the common methods of quantiative forecasting are : 1. 2. 3. 4. 5. Moving Average Weighted Moving Average Exponential Smoothing, Linear Regression Multiple Regression
The appropriateness of method depends on the value to be forecasted. If value to be forecasted is dependent on time. Time series models are useful. If the value to be forecasted is dependent on value of factors other than time and present in Historical data. Causal models such as Linear regression and multiple regression are useful.
2. Mean Forecast Error : Mean forecast error is average of forecast errors for the n observations taken.
Ideal value MFE > 0, model MFE < 0, model tends to over-forecast 3. Mean Absolute Deviation (MAD) :
tends
= to
0; under-forecast
While MFE is a measure of forecast model bias, MAD indicates the absolute size of the errors
Illustration Period 3 4 5 6 7 8 Demand Forecast Error 11 9 10 8 14 12 13.5 13 10 9.5 9 11 -2.5 -4 0 -1.5 5 1 -2 Absolute Error 2.5 4 0 1.5 5 1 14
Number of observations = 6 Mean forecast error = -2/6 = -0.33 Mean Absolut Deviation = 14/6 = 2.33 Conclusion: Model tends to slightly over-forecast, with an average absolute error of 2.33 units.
Evaluating Whether the model being used gives Good forecast or Not ?
A forecast model is good enough OR not can be determined by computing the tracking signal.
Tracking Signal
A forecast can be called good in statistical terms if In Statistical Process Control, people study when a process is going out of control and needs intervention. Tracking signal checks if the forecast extent to which under predicts or over predicts demand because in that case even after keeping some buffer for variation in forecast the firm would find its warehouses increasingly falling short or exceeding capacity. If Tracking signal is above 3.75 or less than -3.75 the system is consistently overforecasting or underforecasting. Illustration :
Why this figure 3.75? At a promised service level of 99% (i.e. 3 sigma level) the error rate is magnitude is 3.75 times Mean Absolute Deviation. Thus Tracking level should be between +3.75 and -3.75 for a good forecast.
Description: 1. Month and Demand : Month and Demand column give the actual value of forecast for the month.
2. 3 month MA :The forecasting is done by taking past 3 months moving average (as shown in third column) 3. Error : The Error column gives the error made by forecast given by the formula : Error = Actual Demand 3 month Moving Average 4. Absolute Error : The Absolute error column gives the magnitude of error. It is used in calculation of Mean Absolute Deviation. 5. Mean Forecast error : The Mean Forecast error gives the direction of forecasting. It is given by the formula : Mean Forecast error = Sum of errors / Number of observation If the value is negative the model overforecasts and if its positive the model underforecasts. In our case Mean Forecast error = (-19 + -7 + 39 + 13 + 8 + -30 + -10 + -12 + -1) / 9 = -2 Thus the model tends to over-forecast. 6. Mean Absolute Deviation : Mean Absolute Deviation = (19 + 7 + 39 + 13 + 8 + 30 + 10 + 12 + 1) / 9 = 15 Thus the average size of overcast is 15 units.
References
http://www.logisticsmgmt.com/article/supply_chain_technology_will_wms_take_over_the_world/ accessed
on 11/12/2012 11:00 PM
http://www.shmula.com/introduction-to-forecasting/307/ accessed on 11/12/2012 12:15 PM
http://scm.ncsu.edu/scm-articles/article/measuring-forecast-accuracy-approaches-to-forecasting-a-tutorial