Vous êtes sur la page 1sur 22

PERT chart (Program Evaluation Review Technique) A PERT chart is a project management tool used to schedule, organize, and

coordinate tasks within a project. PERT stands for Program Evaluation Review Technique , a methodology developed y the !.". #avy in the $%&'s to manage the Polaris su marine missile program. A similar methodology, the Critical Path Method ()P*+ was developed for project management in the private sector at a out the same time. A PERT chart presents a graphic illustration of a project as a network diagram consisting of num ered nodes (either circles or rectangles+ representing events, or milestones in the project linked y la elled vectors (directional lines+ representing tasks in the project. The direction of the arrows on the lines indicates the se,uence of tasks. -n the diagram, for e.ample, the tasks etween nodes $, /, 0, 1, and $' must e completed in se,uence. These are called dependent or serial tasks. The tasks etween nodes $ and /, and nodes $ and 2 are not dependent on the completion of one to start the other and can e undertaken simultaneously. These tasks are called parallel or concurrent tasks. Tasks that must e completed in se,uence ut that don3t re,uire resources or completion time are considered to have event dependency. These are represented y dotted lines with arrows and are called dummy activities. 4or e.ample, the dashed arrow linking nodes 5 and % indicates that the system files must e converted efore the user test can take place, ut that the resources and time re,uired to prepare for the user test (writing the user manual and user training+ are on another path. #um ers on the opposite sides of the vectors indicate the time allotted for the task. The PERT chart is sometimes preferred over the 6antt chart, another popular project management charting method, ecause it clearly illustrates task dependencies. 7n the other hand, the PERT chart can e much more difficult to interpret, especially on comple. projects. 4re,uently, project managers use oth techni,ues.

8ere are some additional resources for learning a out how PERT charts and other project management tools are used in the enterprise9 Project management tools and strategies9 The 6antt chart and the PERT chart are pro a ly the two est known charts in project management. Each of these can e used for scheduling, ut ecause 6antt charts don3t illustrate task dependencies and PERT charts can e confusing, P*s often use oth. Project management charts9 :eyond 6antt9 6antt charts are good for certain purposes, e.plains project management e.pert ;avid )hristiansen, ut there are other charts P*s have at their disposal.

PERT )omple. projects re,uire a series of activities, some of which must e performed se,uentially and others that can e performed in parallel with other activities. This collection of series and parallel tasks can e modeled as a network. -n $%&< the )ritical Path *ethod ()P*+ was developed as a network model for project management. )P* is a deterministic method that uses a fi.ed time estimate for each activity. =hile )P* is easy to understand and use, it does not consider the time variations that can have a great impact on the completion time of a comple. project. The Program Evaluation and Review Technique (PERT+ is a network model that allows for randomness in activity completion times. PERT was developed in the late $%&'3s for the !.". #avy3s Polaris project having thousands of contractors. -t has the potential to reduce oth the time and cost re,uired to complete a project. The Network Diagram -n a project, an activity is a task that must e performed and an event is a milestone marking the completion of one or more activities. :efore an activity can egin, all of its predecessor activities must e completed. Project network models represent activities and milestones y arcs and nodes. PERT originally was an activity on arcnetwork, in which the activities are represented on the lines and milestones on the nodes. 7ver time, some people egan to use PERT as an activity on node network. 4or this discussion, we will use the original form of activity on arc. The PERT chart may have multiple pages with many su >tasks. The following is a very simple e.ample of a PERT diagram9 PERT Chart

The milestones generally are num ered so that the ending node of an activity has a higher num er than the eginning node. -ncrementing the num ers y $' allows for new ones to e inserted without modifying the num ering of the entire diagram. The activities in the a ove diagram are la eled with letters along with the e.pected time re,uired to complete the activity.

Steps in the PERT Planning Process PERT planning involves the following steps9 $. -dentify the specific activities and milestones. /. ;etermine the proper se,uence of the activities. 2. )onstruct a network diagram. 0. Estimate the time re,uired for each activity.
5. ;etermine the critical path.

5. !pdate the PERT chart as the project progresses.

1 !"enti#$ %ctivities an" &ilestones The activities are the tasks re,uired to complete the project. The milestones are the events marking the eginning and end of one or more activities. -t is helpful to list the tasks in a ta le that in later steps can e e.panded to include information on se,uence and duration. ' Determine %ctivit$ Sequence This step may e com ined with the activity identification step since the activity se,uence is evident for some tasks. 7ther tasks may re,uire more analysis to determine the e.act order in which they must e performed. ( Construct the Network Diagram !sing the activity se,uence information, a network diagram can e drawn showing the se,uence of the serial and parallel activities. 4or the original activity>on>arc model, the activities are depicted y arrowed lines and milestones are depicted y circles or ? u les?. -f done manually, several drafts may e re,uired to correctly portray the relationships among activities. "oftware packages simplify this step y automatically converting ta ular activity information into a network diagram. ) Estimate %ctivit$ Times =eeks are a commonly used unit of time for activity completion, ut any consistent unit of time can e used.

A distinguishing feature of PERT is its a ility to deal with uncertainty in activity completion times. 4or each activity, the model usually includes three time estimates9

Optimistic time > generally the shortest time in which the activity can e completed. -t is common practice to specify optimistic times to e three standard deviations from the mean so that there is appro.imately a $@ chance that the activity will e completed within the optimistic time. Most likely time > the completion time having the highest pro a ility. #ote that this time is different from the expected time. Pessimistic time > the longest time that an activity might re,uire. Three standard deviations from the mean is commonly used for the pessimistic time.

PERT assumes a eta pro a ility distri ution for the time estimates. 4or a eta distri ution, the e.pected time for each activity can e appro.imated using the following weighted average9 E.pected time A ( 7ptimistic B 0 . *ost likely B Pessimistic + C 5 This e.pected time may e displayed on the network diagram. To calculate the variance for each activity completion time, if three standard deviation times were selected for the optimistic and pessimistic times, then there are si. standard deviations etween them, so the variance is given y9 D ( Pessimistic > 7ptimistic + C 5 E/

* Determine the Critical Path The critical path is determined y adding the times for the activities in each se,uence and determining the longest path in the project. The critical path determines the total calendar time re,uired for the project. -f activities outside the critical path speed up or slow down (within limits+, the total project time does not change. The amount of time that a non>critical path activity can e delayed without delaying the project is referred to as slack time. -f the critical path is not immediately o vious, it may following four ,uantities for each activity9

e helpful to determine the

E" > Earliest "tart time E4 > Earliest 4inish time F" > Fatest "tart time

F4 > Fatest 4inish time

These times are calculated using the e.pected time for the relevant activities. The earliest start and finish times of each activity are determined y working forward through the network and determining the earliest time at which an activity can start and finish considering its predecessor activities. The latest start and finish times are the latest times that an activity can start and finish without delaying the project. F" and F4 are found y working ackward through the network. The difference in the latest and earliest finish of each activity is that activity3s slack. The critical path then is the path through the network in which none of the activities have slack. The variance in the project completion time can e calculated y summing the variances in the completion times of the activities in the critical path. 6iven this variance, one can calculate the pro a ility that the project will e completed y a certain date assuming a normal pro a ility distri ution for the critical path. The normal distri ution assumption holds if the num er of activities in the path is large enough for the central limit theorem to e applied. "ince the critical path determines the completion date of the project, the project can e accelerated y adding the resources re,uired to decrease the time for the activities in the critical path. "uch a shortening of the project sometimes is referred to as project crashing. + ,p"ate as Pro-ect Progresses *ake adjustments in the PERT chart as the project progresses. As the project unfolds, the estimated times can e replaced with actual times. -n cases where there are delays, additional resources may e needed to stay on schedule and the PERT chart may e modified to reflect the new situation. .ene#its o# PERT PERT is useful ecause it provides the following information9

E.pected project completion time. Pro a ility of completion efore a specified date. The critical path activities that directly impact the completion time. The activities that have slack time and that can lend resources to critical path activities. Activity start and end dates.


The following are some of PERT3s weaknesses9

The activity time estimates are somewhat su jective and depend on judgement. -n cases where there is little e.perience in performing an activity, the num ers may e only a guess. -n other cases, if the person or group performing the activity estimates the time there may e ias in the estimate. Even if the activity times are well>estimated, PERT assumes a eta distri ution for these time estimates, ut the actual distri ution may e different. Even if the eta distri ution assumption holds, PERT assumes that the pro a ility distri ution of the project completion time is the same as the that of the critical path. :ecause other paths can ecome the critical path if their associated activities are delayed, PERT consistently underestimates the e.pected project completion time.

The underestimation of the project completion time due to alternate paths ecoming critical is perhaps the most serious of these issues. To overcome this limitation, *onte )arlo simulations can e performed on the network to eliminate this optimistic ias in the e.pected project completion time.

)ritical Path Analysis and PERT )harts Planning G scheduling more comple. projects Related variants: AOA or Activity-on-Arc Diagrams

*ultiple activities often feed into other activities. )ritical Path Analysis and PERT are powerful tools that help you to schedule and manage comple. projects. They were developed in the $%&'s to control large defense projects, and have een used routinely since then. As with 0antt Charts, )ritical Path Analysis ()PA+ or the )ritical Path *ethod ()P*+ helps you to plan all tasks that must e completed as part of a project. They act as the asis oth for preparation of a schedule, and of resource planning. ;uring management

of a project, they allow you to monitor achievement of project goals. They help you to see where remedial action needs to e taken to get a project ack on course. =ithin a project it is likely that you will display your final project plan as a 6antt )hart (using *icrosoft Project or other software for projects of medium comple.ity or an e.cel spreadsheet for projects of low comple.ity+.The enefit of using )PA within the planning process is to help you develop and test your plan to ensure that it is ro ust. )ritical Path Analysis formally identifies tasks which must e completed on time for the whole project to e completed on time. -t also identifies which tasks can e delayed if resource needs to e reallocated to catch up on missed or overrunning tasks. The disadvantage of )PA, if you use it as the techni,ue y which your project plans are communicated and managed against, is that the relation of tasks to time is not as immediately o vious as with 6antt )harts. This can make them more difficult to understand. A further enefit of )ritical Path Analysis is that it helps you to identify the minimum length of time needed to complete a project. =here you need to run an accelerated project, it helps you to identify which project steps you should accelerate to complete the project within the availa le time. 8ow to !se the Tool9 As with 6antt )harts, the essential concept ehind )ritical Path Analysis is that you cannot start some activities until others are finished. These activities need to e completed in a se,uence, with each stage eing more>or>less completed efore the ne.t stage can egin. These are 3se,uential3 activities. 7ther activities are not dependent on completion of any other tasks. Hou can do these at any time efore or after a particular stage is reached. These are non>dependent or 3parallel3 tasks. ;rawing a )ritical Path Analysis )hart !se the following steps to draw a )PA )hart9 "tep $. Fist all activities in the plan 4or each activity, show the earliest start date, estimated length of time it will take, and whether it is parallel or se,uential. -f tasks are se,uential, show which stage they depend on. 4or the project e.ample used here, you will end up with the same task list as e.plained in the article on 6antt )harts (we will use the same e.ample as with 6antt )harts to compare the two techni,ues+. The chart is repeated in 4igure $ elow9 1igure 1 Task /ist2 Planning a custom3written computer pro-ect Task Earliest /ength T$pe Depen"ent start on A. 8igh level analysis =eek ' $ week "e,uential :. "election of hardware platform ). -nstallation and commissioning of hardware =eek $ $ day =eek $./ "e,uential A :

/ weeksParallel

;. ;etailed analysis of core modules

=eek $ / weeks"e,uential A

E. ;etailed analysis of =eek 2 / weeks"e,uential ; supporting modules 4. Programming of core modules 6. Programming of supporting modules =eek 2 / weeks"e,uential ; =eek & 2 weeks"e,uential E

8. Iuality assurance =eek & $ week "e,uential 4 of core modules -. Iuality assurance =eek 1 $ week "e,uential 6 of supporting modules J.)ore module training K. ;evelopment and IA of accounting reporting F. ;evelopment and IA of management reporting *. ;evelopment of *anagement -nformation "ystem #. ;etailed training =eek 5 $ day Parallel ),8 E

=eek & $ week Parallel

=eek & $ week Parallel

=eek 5 $ week "e,uential F

=eek % $ week "e,uential -, J, K, *

"tep /. Plot the activities as a circle and arrow diagram )ritical Path Analyses are presented using circle and arrow diagrams. -n these, circles show events within the project, such as the start and finish of tasks. The num er shown in the left hand half of the circle allows you to identify each one easily. )ircles are sometimes known as nodes. An arrow running etween two event circles shows the activity needed to complete that task. A description of the task is written underneath the arrow. The length of the task is shown a ove it. :y convention, all arrows run left to right. Arrows are also sometimes called arcs. An e.ample of a very simple diagram is shown elow9

This shows the start event (circle $+, and the completion of the 38igh Fevel Analysis3 task (circle /+. The arrow etween them shows the activity of carrying out the 8igh Fevel Analysis. This activity should take $ week. =here one activity cannot start until another has een completed, we start the arrow for the dependent activity at the completion event circle of the previous activity. An e.ample of this is shown elow9

8ere the activities of 3"elect 8ardware3 and 3)ore *odule Analysis3 cannot e started until 38igh Fevel Analysis3 has een completed. This diagram also rings out a num er of other important points9 =ithin )ritical Path Analysis, we refer to activities y the num ers in the circles at each end. 4or e.ample, the task 3)ore *odule Analysis3 would e called activity / to 2. 3"elect 8ardware3 would e activity / to %. Activities are not drawn to scale. -n the diagram a ove, activities are $ week long, / weeks long, and $ day long. Arrows in this case are all the same length. -n the e.ample a ove, you can see a second num er in the top, right hand ,uadrant of each circle. This shows the earliest start time for the following activity. -t is conventional to start at '. 8ere units are whole weeks. A different case is shown elow9

8ere activity 5 to < cannot start until the other four activities ($$ to 5, & to 5, 0 to 5, and 1 to 5+ have een completed. )lick the link elow for the full circle and arrow diagram for the computer project we are using as an e.ample. 1igure *2 1ull Critical Path Diagram This shows all the activities that will take place as part of the project. #otice that each event circle also has a figure in the ottom, right hand ,uadrant. This shows the latest finish time that3s permissi le for the preceeding activity if the project is to e completed in the minimum time possi le. Hou can calculate this y starting at the last event and working ackwards.The latest finish time of the preceeding event and the earliest start time of the following even will e the same for ciircles on the critical path. Hou can see that event * can start any time etween weeks 5 and 1. The timing of this event is not critical. Events $ to /, / to 2, 2 to 0, 0 to &, & to 5 and 5 to < must e started and completed on time if the project is to e completed in $' weeks. This is the 3critical path3 L these activities must e very closely managed to ensure that activities are completed on time. -f jo s on the critical path slip, immediate action should e taken to get the project ack on schedule. 7therwise completion of the whole project will slip. 3)rash Action3 Hou may find that you need to complete a project earlier than your )ritical Path Analysis says is possi le. -n this case you need to re>plan your project. Hou have a num er of options and would need to assess the impact of each on the projectMs cost, ,uality and time re,uired to complete it. 4or e.ample, you could increase resource availa le for each project activity to ring down time spent on each ut the impact of some of this would e insignificant and a more efficient way of doing this would e to look only at activities on the critical path. As an e.ample, it may e necessary to complete the computer project in 4igure & in 1 weeks rather than $' weeks. -n this case you could look at using two analysts in activities / to 2 and 2 to 0. This would shorten the project y two weeks, ut may raise the project cost L dou ling resources at any stage may only improve productivity y,

say, &'@ as additional time may need to e spent getting the team mem ers up to speed on what is re,uired, coordinating tasks split etween them, integrating their contri utions etc. -n some situations, shortening the original critical path of a project can lead to a different series of activities ecoming the critical path. 4or e.ample, if activity 0 to & were reduced to $ week, activities 0 to 1 and 1 to 5 would come onto the critical path. As with 6antt )harts, in practice project managers use software tools like &icroso#t Pro-ect to create )PA )harts. #ot only do these make them easier to draw, they also make modification of plans easier and provide facilities for monitoring progress against plans. PERT (Program Evaluation and Review Techni,ue+ PERT is a variation on )ritical Path Analysis that takes a slightly more skeptical view of time estimates made for each project stage. To use it, estimate the shortest possi le time each activity will take, the most likely length of time, and the longest time that might e taken if the activity takes longer than e.pected. !se the formula elow to calculate the time to use for each project stage9 shortest time B 0 . likely time B longest time >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 5 This helps to ias time estimates away from the unrealistically short time>scales normally assumed. Key Points9 )ritical Path Analysis is an effective and powerful method of assessing9 =hat tasks must e carried out. =here parallel activity can e performed. The shortest time in which you can complete a project. Resources needed to e.ecute a project. The se,uence of activities, scheduling and timings involved. Task priorities. The most efficient way of shortening time on urgent projects. An effective )ritical Path Analysis can make the difference etween success and failure on comple. projects. -t can e very useful for assessing the importance of pro lems faced during the implementation of the plan. PERT is a variant of )ritical Path Analysis that takes a more skeptical view of the time needed to complete each project stage. 8-"T7RH Program Evaluation an" Review Technique The #avy3s "pecial Projects 7ffice, charged with developing the Polaris>"u marine weapon system and the 4leet :allistic *issile capa ility, has developed a statistical techni,ue for measuring and forecasting progress in research and development

programs. This Program Evaluation and Review Techni,ue (code>named PERT+ is applied as a decision>making tool designed to save time in achieving end>o jectives, and is of particular interest to those engaged in research and development programs for which time is a critical factor. The new techni,ue takes recognition of three factors that influence successful achievement of research and development program o jectives9 time, resources, and technical performance specifications. PERT employs time as the varia le that reflects planned resource>applications and performance specifications. =ith units of time as a common denominator, PERT ,uantifies knowledge a out the uncertainties involved in developmental programs re,uiring effort at the edge of, or eyond, current knowledge of the su ject > effort for which little or no previous e.perience e.ists. Through an electronic computer, the PERT techni,ue processes data representing the major, finite accomplishments (events+ essential to achieve end>o jectivesN the inter>dependence of those eventsN and estimates of time and range of time necessary to complete each activity etween two successive events. "uch time e.pectations include estimates of ?most likely time?, ?optimistic time?, and ?pessimistic time? for each activity. The techni,ue is a management control tool that sizes up the outlook for meeting o jectives on timeN highlights danger signals re,uiring management decisionsN reveals and defines oth criticalness and slack in the flow plan or the network of se,uential activities that must e performed to meet o jectivesN compares current e.pectations with scheduled completion dates and computes the pro a ility for meeting scheduled datesN and simulates the effects of options for decision > efore decision. The concept of PERT was developed y an operations research team staffed with representatives from the 7perations Research ;epartment of :ooz, Allen and 8amiltonN the Evaluation 7ffice of the Fockheed *issile "ystems ;ivisionN and the Program Evaluation :ranch, "pecial Projects 7ffice, of the ;epartment of the #avy. O illard !a"ar #$ead% Program Evaluation &ranch% 'pecial Projects O((ice% )* '* +avy,% The -merican 'tatistician% -pril ./0/ 4verview PERT is a method to analyze the involved tasks in completing a given project, especially the time needed to complete each task, and to identify the minimum time needed to complete the total project. PERT was developed primarily to simplify the planning and scheduling of large and comple. projects. -t was developed for the !.". #avy "pecial Projects 7ffice in $%&< to support the !.". #avy3s Polaris nuclear su marine project. D/E -t was a le to incorporate uncertainty y making it possi le to schedule a project while not knowing precisely the details and durations of all the activities. -t is more of an event>oriented techni,ue rather than start> and completion>oriented, and is used more in projects where time, rather than cost, is the major factor. -t is applied to very large>scale, one>time, comple., non> routine infrastructure and Research and ;evelopment projects. An e.ample of this was for the $%51 =inter 7lympics in 6reno le which applied PERT from $%5& until the opening of the $%51 6ames.D2E This project model was the first of its kind, a revival for scientific management, founded y 4rederick Taylor (Taylorism+ and later refined y 8enry 4ord

(4ordism+. ;uPont3scritical path method was invented at roughly the same time as PERT. Conventions

A PERT chart is a tool that facilitates decision making. The first draft of a PERT chart will num er its events se,uentially in $'s ($', /', 2', etc.+ to allow the later insertion of additional events. Two consecutive events in a PERT chart are linked y activities, which are conventionally represented as arrows (see the diagram a ove+. The events are presented in a logical se,uence and no activity can commence until its immediately preceding event is completed. The planner decides which milestones should e PERT events and also decides their PproperQ se,uence. A PERT chart may have multiple pages with many su >tasks.

PERT is valua le to manage where multiple tasks are occurring simultaneously to reduce redundancy Terminolog$

PERT event9 a point that marks the start or completion of one or more activities. -t consumes no time and uses no resources. =hen it marks the completion of one or more tasks, it is not PreachedQ (does not occur+ until all of the activities leading to that event have een completed. predecessor event9 an event that immediately precedes some other event without any other events intervening. An event can have multiple predecessor events and can e the predecessor of multiple events. successor event9 an event that immediately follows some other event without any other intervening events. An event can have multiple successor events and can e the successor of multiple events. PERT activity9 the actual performance of a task which consumes time and re,uires resources (such as la or, materials, space, machinery+. -t can e understood as representing the time, effort, and resources re,uired to move from one event to another. A PERT activity cannot e performed until the predecessor event has occurred. optimistic time (7+9 the minimum possi le time re,uired to accomplish a task, assuming everything proceeds etter than is normally e.pected pessimistic time (P+9 the ma.imum possi le time re,uired to accomplish a task, assuming everything goes wrong ( ut e.cluding major catastrophes+.

most likely time (*+9 the est estimate of the time re,uired to accomplish a task, assuming everything proceeds as normal. expected time (TE+9 the est estimate of the time re,uired to accomplish a task, accounting for the fact that things don3t always proceed as normal (the implication eing that the e.pected time is the average time the task would re,uire if the task were repeated on a num er of occasions over an e.tended period of time+. TE A (4 B )& B P+ R + (loat or slack is a measure of the e.cess time and resources availa le to complete a task. -t is the amount of time that a project task can e delayed without causing a delay in any su se,uent tasks ( (ree (loat+ or the whole project (total (loat+. Positive slack would indicate ahead o( scheduleN negative slack would indicate 1ehind scheduleN and zero slack would indicate on schedule. critical path9 the longest possi le continuous pathway taken from the initial event to the terminal event. -t determines the total calendar time re,uired for the projectN and, therefore, any time delays along the critical path will delay the reaching of the terminal event y at least the same amount.

critical activity9 An activity that has total float e,ual to zero. An activity with zero float is not necessarily on the critical path since its path may not e the longest. 2ead time9 the time y which a predecessor event must e completed in order to allow sufficient time for the activities that must elapse efore a specific PERT event reaches completion. lag time9 the earliest time y which a successor event can follow a specific PERT event. (ast tracking9 performing more critical activities in parallel crashing critical path9 "hortening duration of critical activities

!mplementation The first step to scheduling the project is to determine the tasks that the project re,uires and the order in which they must e completed. The order may e easy to record for some tasks (e*g* =hen uilding a house, the land must e graded efore the foundation can e laid+ while difficult for others (There are two areas that need to e graded, ut there are only enough ulldozers to do one+. Additionally, the time estimates usually reflect the normal, non>rushed time. *any times, the time re,uired to e.ecute the task can e reduced for an additional cost or a reduction in the ,uality. -n the following e.ample there are seven tasks, la eled - through 3. "ome tasks can e done concurrently (- and &+ while others cannot e done until their predecessor task is complete (C cannot egin until - is complete+. Additionally, each task has three time estimates9 the optimistic time estimate ( O+, the most likely

or normal time estimate (M+, and the pessimistic time estimate (P+. The e.pected time (TE+ is computed using the formula (O B 0M B P+ R 5. Time estimates %ctivit Pre"ecessor $ E5pecte" time 4pt (O) / Normal (M) Pess (P)


& C

O -

2 0

& &

% <

&.22 &.$<

4 E

&, C

0 0

5 &

$' <

5.22 &.$<

! 3

4 E

2 2

0 &

1 1

0.&' &.$<

7nce this step is complete, one can draw a 6antt chart or a network diagram.

A 6antt chart created using *icrosoft Project (*"P+. #ote ($+ the critical path is in red, (/+ the slack is the lack lines connected to non>critical activities, (2+ since "aturday and "unday are not work days and are thus e.cluded from the schedule, some ars on the 6antt chart are longer if they cut through a weekend.

A 6antt chart created using 7mniPlan. #ote ($+ the critical path is highlighted, (/+ the slack is not specifically indicated on task & (d+, though it can e o served on tasks 2 and < ( and f+, (2+ since weekends are indicated y a thin vertical line, and take up no additional space on the work calendar, ars on the 6antt chart are not longer or shorter when they do or don3t carry over a weekend. A network diagram can e created y hand or y using diagram software. There are two types of network diagrams, activity on arrow (A7A+ and activity on node (A7#+. Activity on node diagrams are generally easier to create and interpret. To create an A7# diagram, it is recommended ( ut not re,uired+ to start with a node named start. This ?activity? has a duration of zero ('+. Then you draw each activity that does not have a predecessor activity (a and 1 in this e.ample+ and connect them with an arrow from start to each node. #e.t, since oth c and d list a as a predecessor activity, their nodes are drawn with arrows coming from a. Activity e is listed with 1 and c as predecessor activities, so node e is drawn with arrows coming from oth 1 and c, signifying that e cannot egin until oth 1 and c have een completed. Activity ( has d as a predecessor activity, so an arrow is drawn connecting the activities. Fikewise, an arrow is drawn from e to g. "ince there are no activities that come after ( or g, it is recommended ( ut again not re,uired+ to connect them to a node la eled (inish.

A network diagram created using *icrosoft Project (*"P+. #ote the critical path is in red.

A node like this one (from *icrosoft Sisio+ can name, duration, E", E4, F", F4, and slack.

e used to display the activity

:y itself, the network diagram pictured a ove does not give much more information than a 6antt chartN however, it can e e.panded to display more information. The most common information shown is9 $. The activity name /. The normal duration time 2. The early start time (E"+ 0. The early finish time (E4+ &. The late start time (F"+ 5. The late finish time (F4+ <. The slack -n order to determine this information it is assumed that the activities and normal duration times are given. The first step is to determine the E" and E4. The E" is defined as the ma.imum E4 of all predecessor activities, unless the activity in ,uestion is the first activity, for which the E" is zero ('+. The E4 is the E" plus the task duration (E4 A E" B duration+.

The E" for start is zero since it is the first activity. "ince the duration is zero, the E4 is also zero. This E4 is used as the E" for a and 1. The E" for a is zero. The duration (0 work days+ is added to the E" to get an E4 of four. This E4 is used as the E" for c and d. The E" for 1 is zero. The duration (&.22 work days+ is added to the E" to get an E4 of &.22. The E" for c is four. The duration (&.$< work days+ is added to the E" to get an E4 of %.$<. The E" for d is four. The duration (5.22 work days+ is added to the E" to get an E4 of $'.22. This E4 is used as the E" for (. The E" for e is the greatest E4 of its predecessor activities (1 and c+. "ince 1 has an E4 of &.22 and c has an E4 of %.$<, the E" of e is %.$<. The duration (&.$< work days+ is added to the E" to get an E4 of $0.20. This E4 is used as the E" for g. The E" for ( is $'.22. The duration (0.& work days+ is added to the E" to get an E4 of $0.12. The E" for g is $0.20. The duration (&.$< work days+ is added to the E" to get an E4 of $%.&$. The E" for (inish is the greatest E4 of its predecessor activities (( and g+. "ince ( has an E4 of $0.12 and g has an E4 of $%.&$, the E" of (inish is $%.&$. !inish is a milestone (and therefore has a duration of zero+, so the E4 is also $%.&$.

:arring any unforeseen events, the project should take $%.&$ work days to complete. The ne.t step is to determine the late start (F"+ and late finish (F4+ of each activity. This will eventually show if there are activities that have slack. The F4 is defined as the minimum F" of all successor activities, unless the activity is the last activity, for which the F4 e,uals the E4. The F" is the F4 minus the task duration (F" A F4 > duration+.

The F4 for (inish is e,ual to the E4 ($%.&$ work days+ since it is the last activity in the project. "ince the duration is zero, the F" is also $%.&$ work days. This will e used as the F4 for ( and g. The F4 for g is $%.&$ work days. The duration (&.$< work days+ is su tracted from the F4 to get an F" of $0.20 work days. This will e used as the F4 for e.

The F4 for ( is $%.&$ work days. The duration (0.& work days+ is su tracted from the F4 to get an F" of $&.'$ work days. This will e used as the F4 for d. The F4 for e is $0.20 work days. The duration (&.$< work days+ is su tracted from the F4 to get an F" of %.$< work days. This will e used as the F4 for 1 and c. The F4 for d is $&.'$ work days. The duration (5.22 work days+ is su tracted from the F4 to get an F" of 1.51 work days. The F4 for c is %.$< work days. The duration (&.$< work days+ is su tracted from the F4 to get an F" of 0 work days. The F4 for 1 is %.$< work days. The duration (&.22 work days+ is su tracted from the F4 to get an F" of 2.10 work days. The F4 for a is the minimum F" of its successor activities. "ince c has an F" of 0 work days and d has an F" of 1.51 work days, the F4 for a is 0 work days. The duration (0 work days+ is su tracted from the F4 to get an F" of ' work days. The F4 for start is the minimum F" of its successor activities. "ince a has an F" of ' work days and 1 has an F" of 2.10 work days, the F" is ' work days.

The ne.t step is to determine the critical path and if any activities have slack. The critical path is the path that takes the longest to complete. To determine the path times, add the task durations for all availa le paths. Activities that have slack can e delayed without changing the overall time of the project. "lack is computed in one of two ways, slack A F4 > E4 or slack A F" > E". Activities that are on the critical path have a slack of zero ('+.

The duration of path ad( is $0.12 work days. The duration of path aceg is $%.&$ work days. The duration of path 1eg is $&.5< work days.

The critical path is aceg and the critical time is $%.&$ work days. -t is important to note that there can e more than one critical path (in a project more comple. than this e.ample+ or that the critical path can change. 4or e.ample, let3s say that activities d and ( take their pessimistic ( + times to complete instead of their e.pected (T E+ times. The critical path is now ad( and the critical time is // work days. 7n the other hand, if activity c can e reduced to one work day, the path time for aceg is reduced to $&.20 work days, which is slightly less than the time of the new critical path, 1eg ($&.5< work days+.

Assuming these scenarios do not happen, the slack for each activity can now e determined.

'tart and (inish are milestones and y definition have no duration, therefore they can have no slack (' work days+. The activities on the critical path y definition have a slack of zeroN however, it is always a good idea to check the math anyway when drawing y hand.

F4a > E4a A 0 > 0 A ' F4c > E4c A %.$< > %.$< A ' F4e > E4e A $0.20 > $0.20 A ' F4g > E4g A $%.&$ > $%.&$ A '

Activity 1 has an F4 of %.$< and an E4 of &.22, so the slack is 2.10 work days. Activity d has an F4 of $&.'$ and an E4 of $'.22, so the slack is 0.51 work days. Activity ( has an F4 of $%.&$ and an E4 of $0.12, so the slack is 0.51 work days.

Therefore, activity 1 can e delayed almost 0 work days without delaying the project. Fikewise, activity d or activity ( can e delayed 0.51 work days without delaying the project (alternatively, d and ( can e delayed /.20 work days each+.

A completed network diagram created using *icrosoft Sisio. #ote the critical path is in red. %"vantages

PERT chart e.plicitly defines and makes visi le dependencies (precedence relationships+ etween the =:" elements PERT facilitates identification of the critical path and makes this visi le PERT facilitates identification of early start, late start, and slack for each activity, PERT provides for potentially reduced project duration due to etter understanding of dependencies leading to improved overlapping of activities and tasks where feasi le. The large amount of project data can e organized G presented in diagram for use in decision making.


There can e potentially hundreds or thousands of activities and individual dependency relationships

PERT is not easily scala le for smaller projects The network charts tend to e large and unwieldy re,uiring several pages to print and re,uiring special size paper The lack of a timeframe on most PERTC)P* charts makes it harder to show status although colours can help (e.g., specific colour for completed nodes+ =hen the PERTC)P* charts ecome unwieldy, they are no longer used to manage the project.

,ncertaint$ in pro-ect sche"uling ;uring project e.ecution, however, a real>life project will never e.ecute e.actly as it was planned due to uncertainty. -t can e am iguity resulting from su jective estimates that are prone to human errors or it can e varia ility arising from une.pected events or risks. The main reason that the Program Evaluation and Review Techni,ue (PERT+ may provide inaccurate information a out the project completion time is due to this schedule uncertainty. This inaccuracy is large enough to render such estimates as not helpful. 7ne possi ility to ma.imize solution ro ustness is to include safety in the aseline schedule in order to a sor the anticipated disruptions. This is called proactive scheduling. A pure proactive scheduling is a utopia, incorporating safety in a aseline schedule that allows to cope with every possi le disruption would lead to a aseline schedule with a very large make>span. A second approach, reactive scheduling, consists of defining a procedure to react to disruptions that cannot e a sor ed y the aseline schedule.