Bài giảng môn Quản trị kinh doanh - Chapter 8: Timeliness: Scheduling and project management

Chapter 8Timeliness: Scheduling and Project Management1Learning ObjectivesExplain the effect time has on each profitability measure. Describe the impact of feedback delay on quality. Explain the common time reduction strategies. Construct a Gantt chart. Sequence orders using the traditional sequencing rules. Describe the physical features of queues and describe how they affect  queue performance. Compute the probability of x arrivals per unit time for a queue. Describe the psychological approaches to managing perception of queue  length. Construct a network diagram for a project. Identify the critical path for a project using CPM calculations Calculate the likelihood of completing a project in a specified time. Complete the calculations necessary to effectively crash a project.2Time is moneyIt affects customer perceived valueIt affects Net Income It affects Return on AssetsOperations Management Framework3Time and Customer ValueSpeed and flexibilityFaster processes allow greater flexibilityDeliver more options in the same amount of timeCustomers will only wait so long.Start later, closer to demandBetter predictions of what will be needed4Time Really Is MoneyTime affects profitability through net incomeTime-based value attributes - Increasing net salesResponse timeDelivery dependability5Time Really Is MoneyManaging timing of payments to make money (managing “float”)Example. . . .Travel AgentsYou pay right away (on a credit card)They pay in 30 days (business billing cycle) They deep the interestExample . . . . .Insurance companiesCollect only as much as (or less than) they will eventually pay in claimsInvest money while they are waiting to pay out claims6The Effect of Time on ROAThe cash-to-cash cycleNot just production time. . .Waiting for supplies to arrive -- Opportunity cost of salesWaiting to sell product -- Delay of return on investment in inventory and productionWaiting to collect from customer -- Financing the customer’s purchase7The Effect of Time on ROASuppose the transaction in exhibit 8.1 netted a $200 return on total expenditures of $1000 (a 20% return). What would happen to ROA if the cash-to-cash cycle went from 17 weeks to 13 weeks?With a 17 week cash-to-cash cycle, there are approximately three transactions per year for a return of $600 (60%). Reducing the cash-to-cash cycle time to 13 weeks results in the $1000 being turned over four times instead of three. $800 will now be generated. The yield shifts from 60% to 80%.A 23.5% reduction in time results in a 33% increase in return financial return8The Effect of Time on ROAShorter delivery lead-times from suppliers mean lower inventory requirementsWeekly DeliveryAverage Inventory (Cases) = 400 casesAverage Inventory (Investment) = $106,000Daily DeliveryAverage Inventory (Cases) = 80 casesAverage Inventory (Investment) = $21,200 Example from Chapter 2Deliveries of 800 cases weekly (all are consumed)Cases cost $265, Sell for $300Weekly Net Income is $28,000ROA = $28,000/$106,000 = 26.42%ROA = $28,000/$21,200 = 132.08%As inventory is reduced to 1/5, ROA increases five-fold9Feedback Delay. . . Another benefit of time reduction.Elimination of causes of quality problems depends on timely information. The greater the elapsed time between producing a defect and discovering it, the less likely the defects cause will be found.10General Time Reduction StrategiesTwo general approaches:Compressing time of activitiesMaking tasks concurrent instead of sequential 11General Time Reduction StrategiesInsert exhibit 8.312General Time Reduction StrategiesInsert exhibit 8.413General Time Reduction StrategiesInsert exhibit 8.514SchedulingDefinition: Determination of when something is to be done and the tasks and activities required to do itScheduling aids in on-time completionDirect link to value perceived by customersScheduling improves the utilization of the firm’s resourcesDirect link to productivityForward schedulingWhen start date is known and completion needs determinationBackward schedulingWhen completion date is known and start date needs determination15SchedulingGantt ChartsThe layout of steps within a processThe layout of multiple tasks that utilize a single resource.16Traditional Job Shop SequencingSequencing rules for scheduling multiple jobs at a single resourceFirst Come, First ServedEarliest Due DateShortest Processing TimeCritical RatioTime Remaining/effort remainingC.R. < 1 means there is no way to complete the job on time17Example 8.1 Job Sequencing Using EDD18Example 8.1 Job Sequencing Using EDD19Example 8.2 Job Sequencing Using SPT20Example 8.3 Job Sequencing Using Critical Ratio21Traditional Job Shop SequencingMost traditional sequencing rules assume a single operation.Slack Time RemainingSlack = Days to due date - Total processing time (all operations)Slack per remaining operation = Slack / Number of remaining operationsExhibit 8.12: Slack per Remaining Operations22People Are Not Products:People Expect To Be Treated “Fairly”Queue configuration: The physical design of the lines and servers in a queuing system.Server: A resource that is able to complete the process or service that customers or jobs wait in queue for.Phase: A distinct step in a process that requires a separate queue.Jockeying: When customers switch lines hoping to move faster.Queue Discipline: The rules that management enforces to determine the next customer served in a queue.Calling Population: the population of arriving customers or orders.23The Queue Arrival Process: Determining the Probability of x arrivals in a time periodQueue arrivals follow a Poisson distribution.P(x) = the probability of x arrivals in a time periodx = the number of arrivals per unit timeλ = the average arrival rate in a certain time incremente = 2.7283 (the base of the natural logarithms)24Managing Queue ArrivalsA Call center receives 36 calls per hour, Poisson distributed. What is the probability of receiving 41 calls?λ = 36x = 41e = 2.7283 (the base of the natural logarithms)25People Are Not Products:Treating People “Fairly”is ImportantFirst Come, First Served seems the most fair26People Are Not Products:Serving Customers in QueuePsychological features of queuesMake the wait seem less be . . .Keeping customers busyKeeping customers informedTreating customers fairlyStarting the service as soon as possibleExceeding the customer’s expectations27People Are Not Products:Queue AlternativesAvoid Queues: Developing Alternate ProceduresExpress check-in and check-out (hotels, rental cars)Pay bills by mail and online, not in personParking tickets paid onlineAutomatic Checking Account DebitsPhone,Utilities, Credit Cards, Mortgages28Project ManagementA project is a set of activities aimed at meeting a goal, with a defined beginning and end.29Project Management with Certain Time EstimatesSummary of steps:Determine activities that need to be accomplishedDetermine precedence relationships and completion timesConstruct network diagramDetermine the critical pathDetermine early start and late start schedules30Create Network DiagramBased on order of precedence among activitiesProject Management: Scheduling Projects with Certain Time Estimates31Calculation of the Critical PathNetwork approach helps calculate project durationA “path” is a sequence of activities that begins at the start of the project and goes to the end of the project1,2,3,5,6,7,81,2,4,6,7,8The “critical path” is the path that takes the longest to completeand thus determines the minimum duration of the project1243567832Calculation of the Critical PathExhibit 8.17: Project Detail33Calculation of the Critical PathCritical PathThe path that takes the longest to complete124356784 weeks10 weeks4 weeks2 weeks3 weeks16 weeks4 weeks1 weekExhibit 8.17: Project Detail34Calculation of the Critical PathCritical PathThe path that takes the longest to completeC.P. = 40 weeksExhibit 8.17: Project Detail124356784 weeks10 weeks4 weeks2 weeks3 weeks16 weeks4 weeks1 week35Calculation of the Critical PathIt is possible for multiple Critical Paths to existSuppose new information suggested that Activity 4 would take 5 weeks instead of 4. . .X5Exhibit 8.17: Project Detail124356784 weeks10 weeks4 weeks2 weeks3 weeks16 weeks4 weeks1 week36Calculation of the Critical PathX5Exhibit 8.17: Project Detail124356784 weeks10 weeks5 weeks2 weeks3 weeks16 weeks4 weeks1 weekThere are now two critical paths, each one is 40 weeks.37Calculation of the Critical PathThe Critical Path may also shift if non-critical activity is lengthened or Critical Path activity is shortenedSuppose another update indicates that it will actually take 6 weeks for Activity 4The Critical Path becomes 41 days.4 weeks1243567810 weeks6 weeks2 weeks3 weeks16 weeks4 weeks1 weekX6Exhibit 8.17: Project Detail38Determining SlackSlack - The amount of time an activity on a non-critical path can be delayed without affecting the duration of the project (i.e., without putting it on the critical path)Four values are usedEarly start - Earliest an activity can start (based on prerequisites)Early finish - Earliest it can finish (based on prerequisites & duration)Late start - Latest an activity can start and not delay the projectLate finish - Latest an activity can finish and not delay the project39Calculating Early Start (ES) and Early Finish (EF)The process moves from left to right in network The ES for the 1st activity is usually zeroThe EF equals the ES plus activity duration ES is the latest of the EF times  of an activity’s predecessorsExhibit 8.17: Project Detail40This process moves from right to left in the networkCalculating Late Start (LS) and Late Finish (LF) LF for the last activity equals the EF for the last activity LS equals LF minus the activity durationLF is the earliest of the LS times of an activity’s followers41Calculating SlackSlack - The amount of time an activity on a non-critical path can be delayed without affecting the duration of the project (i.e., without putting it on the critical path)Can be computed by either: Late Start - Early Start or Late Finish - Early FinishActivities that have zero slack form the critical pathExhibit 8.21: ES, EF, LS, LF, and Slack Values42Project Scheduling When the Times of Activities are Uncertain Summary of steps:Determine the activities that need to be accomplishedDetermine the precedence relationships and completion timesConstruct the network diagramDetermine the critical pathDetermine the early start and late start schedulesCalculate the variances for the activity timesCalculate the probability of completing by the desired due date43Project Scheduling with Time UncertaintyThe “Heuristic” approach to dealing with timing uncertaintyBased on understanding of individual activities as conforming to a “beta” distributionTake three time estimatesOptimistic - What is the (realistic) fastest we can get an activity done?Pessimistic - What is the (realistic) worst case scenario for delay?Most likely - What is our “most likely” estimate?44Project Scheduling with Time UncertaintyCalculate the “expected time” for the activityWhere:t = Expected Timeo = Optimistic estimatem = Most likely estimatep = Worst-case (pessimistic) estimate45Project Scheduling with Time UncertaintyActivity #1weeksActivity #3weeksActivity #2weeks46124356784 weeks10 weeks4.17 weeks2.17 weeks3 weeks16.17 weeks4.17 weeks1.17 weeksC.P. = 40.68 weeksHow do we interpret this estimate???There is a 50% chance we will finish faster, and a 50% chance that we will finish slower than our estimateWe assume that finish times across the entire project are distributed normally.Using Variation in Time Estimates to Assess Duration Probabilities47So, if the “expected” duration (t) of the project is 40.68 weeks, how do we determine the probability of finishing it in 39 weeks (or less)?Take advantage of the estimates of t and make an estimate of 2 draw on knowledge of statistics (especially Normal distributions), and use the “standard normal probability” table along with the following equation:Using Variation in Time Estimates to Access Duration Probabilities48Where:Z = Number of standard deviations D is from Tet = Expected TimeD = Project duration I am thinking aboutNote that this is , not 2 ,We have to take the square root of the “variance” to get the standard deviationUsing Variation in Time Estimates to Determine Duration Probabilities49We recognize that there is variation around our estimatesWe are estimating “most likely” or “expected” not “exact”Commonly assume to be a “Beta” distributionCalculating variance of the activity’s duration estimate:Using Variation in Time Estimates to Access Duration Probabilities50Activity #1weeksActivity #2weeksActivity #3weeksUsing Variation in Time Estimates to Access Duration Probabilities51Using Variation in Time Estimates to Assess Duration ProbabilitiesCalculate Critical Path using t Accumulate the variances of the individual activitiesApply formula to estimate project duration probabilitiesWhere:Z = Number of standard deviations D is from tt = Expected TimeD = Activity duration I am thinking about52Using Variation in Time Estimates to Assess Duration ProbabilitiesExhibit 8.22: Expected Values and Variances of Time Estimates53124356784 weeks10 weeks4.17 weeks2.17 weeks3 weeks16.17 weeks4.17 weeks1.17 weeksUsing Variation in Time Estimates to Assess Duration ProbabilitiesWhat is the probability of finishing in 39 weeks or less?Exhibit 8.22: Expected Values and Variances of Time Estimates54124356784 weeks10 weeks4.17 weeks2.17 weeks3 weeks16.17 weeks4.17 weeks1.17 weeksC.P. = 40.68 weeksSum of Variances = 2.56 weeksUsing Variation in Time Estimates to Assess Duration ProbabilitiesWhat is the probability of finishing in 39 weeks or less?Exhibit 8.22: Expected Values and Variances of Time Estimates55Using Variation in Time Estimates to Assess Duration ProbabilitiesLook up the cumulative probability of the activity being completed before -1.05 standard deviations (Appendix B, near the center)There is a 14.686% chance of finishing in 39 weeks or less. Where:Z = Number of standard deviations D is from tt = Expected TimeD = Activity duration I am thinking aboutUSE:56Crashing ProjectsA methodical approach to reducing project durationFocus on the time of activities on the critical pathLooking for greatest improvement with least costAdditional labor, machineryOvertime and temporary employeesPremiums paid to outside contractors for early deliveryStepsCreate networkIdentify critical pathIdentify costs of reducing each activity on pathReduce most cost effective activityLook for critical path changesBeware of multiple critical pathsCrash next activity57Crashing Projects: Create the NetworkExhibit 8.24: Network to Crash58A8B7D10E12C7F8G9H5Crashing Projects: Identify the Critical PathC.P. = 35 days59Crashing Projects:Identify Costs of Crashing Each ActivityExhibit 8.25: Crash Time and Costs60A8B7D10E12C7F8G9H5Crashing Projects:Reduce Most Cost Effective Activity6XCritical path starts out at 35 days61A8B7D10E12C7F8G9H5Crashing Projects:Look for Critical Path Changes 6XCritical path now= 35 days62A8B7D10E12C7F8G9H5Crashing Projects:Look for Critical Path Changes 6XThere are now 2 critical paths, so reducing the project length will require shortening each63Crashing Projects:Crash Next ActivityBoth C.P.Path 1 OnlyPath 2 OnlyExhibit 8.25: Crash Time and Costs64Crashing Projects: Summary Solution Exhibit 8.26: Crashing Summary65CaveatsTime estimates are frequently wrongEarly finishes are absorbedCushions get wastedResources aren’t always available when needed66