Integrated Cost Schedule Risk Analysis Using the Risk Driver Approach

Transcription

1 Integrated Cost Schedule Risk Analysis Using the Risk Driver Approach Qatar PMI Meeting February 19, 2014 David T. Hulett, Ph.D. Hulett & Associates, LLC 1

2 The Traditional 3-point Estimate of Activity Duration Uncertainty Typical schedule risk analysis starts with the activity that is impacted by risks Estimates the 3-point estimate for optimistic, most likely and pessimistic duration Creates a probability distribution for activity duration Performs Monte Carlo simulation What causes the most overall schedule risk? These questions are answered in the 3-point estimate approach by looking to the impacted activities: Sensitivity to activity durations Criticality of activity durations 2

3 Some Problems with Traditional Approach Can tell which activities are crucial, but not directly which risks are driving Makes poor use of the Risk Register that is usually available Cannot decompose the overall schedule risk into its components BY RISK Ability to assign the risk to its specific risk drivers helps with communication of risk causes and risk mitigation 3

4 Risk Drivers: Start with the Risks Themselves Drive the schedule risk by the risks already analyzed in the Risk Register Root Causes For each risk, specify: Probability it will occur Impact on time if it does Activities it will affect Starting with the risks themselves gives us benefits Links qualitative analysis to the quantitative analysis Estimates the impact of specific risks for prioritized mitigation purposes Correlations between activities happen automatically never have to guess at these coefficients again, never get impossible matrices 4

5 Simple Example of Risk Register Risks Use the Risk Factors feature in Pertmaster 8.7 A few other simulation packages have this capability Collect probability and impact data on risks Load the risks Assign risks to schedule activities 5

6 Risk Factors Mechanics (1) The risk factor is assigned to one or several activities, affecting their durations by a multiplicative factor E.g., the factor may be.90 for optimistic, 1.0 for most likely and 1.25 for pessimistic These factors multiply the schedule durations of the activities to which they are assigned Risks can be assigned to one or more activities Activity durations can be influenced by one or more risks 6

7 Risk Factors Mechanics (2) Risk Factors are assigned a probability of occurring on any iteration. When the risk occurs, the factor used is chosen at random from the 3-point estimate and operates on all activities to which it is assigned When not occurring on an iteration the risk factor takes the value 1.0, a neutral value When an activity is influenced by more than one risk, their factors are multiplied together, if they happen, on any iteration if they are specified to be in series 7

8 Risk Factor Probability is 100%, Factor can be + or - Entire Plan : Duration 100% 115 Here the Ranges are based on deviations + and from the Plan. Probability is 100% Hits % % % % % % % % % % % % % 99 30% 99 Cumulative Frequency For the examples we use an activity with 100 days in the schedule % 98 20% 97 15% % 95 5% Distribution (start of interval) 0% 90 8

9 Assigning a Probability Less than 100% The essence of a risk is its uncertainty in two dimensions: Uncertainty of its occurrence, specified by a probability Uncertainty of its impact, specified by a range of durations If the risk may or may not occur, we specify the probability that it will occur The risk occurs and affects the activities it is assigned to on X% of the iterations, chosen at random, the multiplicative factor used is chosen at random from the range of data input by the user On (1 X)% of the iterations, Factor takes 1.0 value 9

10 Assigning a Probability Less than 100% Spike contains 40% of the probability Hits Entire Plan : Duration 100% % % % % % % % % % % % % % 100 Cumulative Frequency Hits Entire Plan : Duration 100% % % % % % % % % % % % % % 100 Cumulative Frequency Spike contains 70% of the probability % % % % % % % % % % 99 5% 100 5% Distribution (start of interval) 0% Distribution (start of interval) 0% 90 10

11 Assigning More than One Risk to an Activity If more than one risk is acting on an activity, the resulting ranges are the multiplication of the percentages This is reality an activity is often affected by multiple risks Two cases are shown next: When both risks are 100% likely to occur When both risks are < 100% likely to occur In each case, the computer simulation creates the uncertainty range on an activity s duration it is not estimated 11

12 Parallel and Series Risks Additive used with Risk Register Risk 1 20 days Risk 2 12 days 20 days for risk recovery If these two risks are parallel, they can be recovered simultaneously Risk 1 20 days Risk 2 12 days 32 days for risk recovery If these two risks are series, they can not be recovered simultaneously so the duration is longer 12

13 Parallel and Series Risks Multiplicative Used with Risk Drivers (RiskFactors) Risk factor Risk factor Use 1.2 Factor, the largest factor only If these two risks are parallel, they can be recovered simultaneously Risk factor Risk factor If these two risks are series, they can not be recovered simultaneously Use (1.2 x 1.05 = 1.26) Factor, multiply the two 13

14 Two Risks affect One Activity using Factors that Occur 100% - placed in Series Entire Plan : Duration 100% % % % % 123 Risks in series, P80 is 123 days Hits % % % % % % % % % 111 Cumulative Frequency 80 30% % % % % 104 5% Distribution (start of interval) 0% 92 14

15 Two Risks affect One Activity using Factors that Occur 100% - in Parallel Entire Plan : Duration 100% % % % % % % % 116 Risks in parallel, P80 is 119 days Hits % % % % % % % 110 Cumulative Frequency % % % % 106 5% Distribution (start of interval) 0%

16 Two Risks with Less than 100% Probability Affecting one Activity Risks in Series The spike at 100 days represents (1) the likelihood that neither risk occurs [60% x 50% = 30%] plus (2) the chance that 100 days is picked when one or both occur. Hits Entire Plan : Duration 100% % % % % % % % % % % % % % % % % % % 99 5% 96 Cumulative Frequency Distribution (start of interval) 0% 90 16

17 Two Risks with Less than 100% Probability Affecting one Activity Risks in Parallel With one risk s having a minimum range of 100%, it cannot be less than 100 days Hits Entire Plan : Duration 100% % % % % % % % % % % % % 100 Cumulative Frequency % % % % % % 100 5% Distribution (start of interval) 0%

18 Risk Factors Model How Correlation Occurs Coefficients are Calculated (1) Risk 1: Risk Probability =.5, Range.95, 1.05, 1.15 Activity 1 Activity 2 Correlation = 100% 18

19 Scatter showing 100% Correlation 19

20 Risk Factors Model How Correlation Occurs Coefficients are Calculated (2) Risk 3: Risk Probability =.25, Range.8,.95, 1.05 Risk 1: Risk Probability =.5, Range.95, 1.05, 1.15 Risk 2: Risk Probability =.45, Range 1.0, 1.10, 1.20 Activity 1 Activity 2 Correlation = 37% Correlation is modeled as it is caused in the project Correlation coefficients are generated, not guessed 20

21 Scatter showing 37% Correlation 21

22 Sensitivity to the Risk Factors Risk 1 is more important since it affects both Activity A and Activity B 22

23 Integrated Cost and Schedule Risk Analysis Risk Project Schedule Risk Schedule Risk Burn Rate per day Cost Risk Time Independent Costs Time-Dependent Costs Project Cost Risk 23

24 Use a Case Study to Illustrate the Integration of Cost and Schedule Risk Case Study is a simplified schedule of an off-shore gas production project including approval process, drilling, fabrication of the jacket and the topsides (modules), installation of same, and commissioning. Use Risk Drivers approach Results include: Schedule risk and cost risk histograms and cumulative distributions Prioritizing risks to time and to cost Scatter diagram of cost and time Probabilistic cash flow 24

25 How the Impact Ranges work on Different Resources to show Cost Risk (1) The resources are either time dependent or time independent Time Dependent resources such as labor and rented equipment (barges, tower cranes) vary according to: Schedule risk multiplied by the burn rate per day Burn rate is uncertain if the risk exhibits a range So time dependent resources costs vary: By the duration uncertainty even if the burn rate is fixed By the uncertain burn rate even if durations are fixed By both uncertain duration and uncertain burn rate, a cross-product 25

26 How the Impact Ranges work on Different Resources to show Cost Risk (2) Time-independent resources such as procured raw materials and equipment, perhaps some subcontracts The total estimated cost is varied by the range on a risk with its probability Time independent resources costs: May vary even if the duration of Fab and Deliver activities is fixed If they have no cost impact ranges they do not vary even if Fab and Deliver activities durations are variable 26

27 Offshore Gas Production: Integrating Schedule and Cost Risk Analysis 27

28 Resources and Cost Total Project = $1,473 million Resources are added to the activities PMT = Project Management Team Detail = Detailed Engineering PROC = Procurement FAB = Fabrication DRILL = Drilling INST = Installation HUC = Hook-Up and Commissioning We could have LOE costs such as PMT placed on a hammock in PRA 28

29 Inherent Variation and Duration Estimating Uncertainty The estimate is that durations are underestimated The range, on all activities, is 95% 105% - 115% of the estimated duration, indicating some bias in the assumptions and / or estimating Different activity types could have different reference ranges for their inherent variation Inherent variation is not susceptible to risk mitigation it is always with us and must be recognized 29

30 Schedule Risk with Inherent uncertainty Offshore Gas Production Project Entire Plan : Finish Date 100% 05 Aug % 05 Jul 16 90% 28 Jun % 23 Jun 16 Hits % 19 Jun 16 75% 15 Jun 16 70% 13 Jun 16 65% 10 Jun 16 60% 07 Jun 16 55% 04 Jun 16 50% 01 Jun 16 45% 30 May 16 40% 27 May 16 Cumulative Frequency Baseline schedule date is 20MAR16 P-80 is 19JUN % 25 May 16 30% 22 May % 19 May % 16 May % 12 May 16 10% 07 May % 29 Apr Apr Jun Aug 16 Distribution (start of interval) 0% 31 Mar 16 30

31 Cost Risk with Inherent Uncertainty Offshore Gas Production Project Entire Plan : Cost 100% $1,534, % $1,518, % $1,514, % $1,511, Hits % $1,509, % $1,507, % $1,506, % $1,504, % $1,503, % $1,501, % $1,500, % $1,498, % $1,497, Cumulative Frequency Baseline cost is $1,473 million Cost at the P-80 is $1,509 million 35% $1,496, % $1,494, % $1,492, % $1,491, % $1,488, % $1,486, % $1,482, $1,470, $1,480, $1,490, $1,500, $1,510, $1,520, $1,530, Distribution (start of interval) 0% $1,465,

32 Risk Drivers Used Schedule Impact Ranges Cost Impact Ranges Schedule Impact Ranges Cost Impact Ranges Risk Name Min Most Most Max Min Likely Likely Max Probability 1 Company's Engineers may vary in Experience 90% 110% 130% 95% 100% 110% 85% 2 MTO, Specs may not be ready for ITB 100% 105% 125% 100% 105% 110% 60% 3 May have trouble interfacing phases 100% 110% 120% 90% 105% 115% 40% 4 Subsea conditions may not be as expected 95% 100% 115% 95% 100% 110% 20% 5 LLE suppliers may be busy 80% 95% 110% 95% 105% 115% 20% 6 Quality engineers at Fabricators may be inexperienced 100% 100% 100% 100% 105% 110% 60% 7 Scope growth may be greater than expected 105% 110% 125% 100% 105% 110% 60% 8 HUC resources may not be as experienced as needed 100% 115% 130% 95% 105% 115% 20% These data are derived during in-depth interviews with project participants and others. The interviews focus on the Risk Register risks that are designated high risk for time and cost. Use Primavera risk Analysis Risk Factor capability 32

33 Assignment of Risk Drivers to Activities Risk Driver Market Costs for Bulks/Equipment Volatile Experienced HUC resources availability Activity Assignment DETAIL FAB PROC INSTAL HUC DRILL PMT APPROVAL X X X X X Company's Engineers' experience X X X X X X X X Schedule Maturity X X X X X X X MTO, Specifications may not be ready ITB X X X Problems interfacing Phases X X X X Cost Estimate inaccurate / immature X X X X X X X Fabricators and Suppliers may be busy X X Quality engineers may be FAB, Suppliers Scope Growth may be more than expected X X X X 33

34 Some Risks are Parallel and Some Series If several risks are assigned to any task we may put some in parallel (can be recovered simultaneously) and some in series (recovery will take all resources so cannot be recovered simultaneously with other risks) In this case, risks 6 and 7 have been placed in series 34

35 Schedule Risk Analysis Results Schedule Risk Analysis Date Results for First Gas Baseline Date 20-Mar-16 Risk Analysis Results P-5 P-50 P-80 P-95 2-Jul-16 4-Nov-16 6-Jan Mar-17 Months from Baseline Duration Results To First Gas Days Baseline Duration 1,906 Risk Analysis Results P-5 P-50 P-80 P-95 2,010 2,135 2,198 2,265 Percentage from Baseline 5% 12% 15% 19% 35

36 Risks and Inherent Uncertainty Completion Date First Gas First Gas Baseline Date = 20MAR Offshore Gas Production Project First Gas : Finish Date 100% 19 Jul 17 95% 14 Mar 17 90% 14 Feb 17 85% 22 Jan 17 80% 06 Jan 17 Date with Inherent uncertainty was 19JUN16 P-80 with all risks added is 6Jan months Hits % 24 Dec 16 70% 11 Dec 16 65% 30 Nov 16 60% 20 Nov 16 55% 13 Nov 16 50% 04 Nov 16 45% 26 Oct 16 40% 17 Oct 16 35% 08 Oct 16 30% 29 Sep 16 25% 18 Sep 16 20% 06 Sep 16 15% 22 Aug 16 10% 31 Jul 16 Cumulative Frequency 5% 02 Jul Apr Aug Nov Feb Jun 17 Distribution (start of interval) 0% 22 Apr 16 36

37 Risks and Inherent Uncertainty Schedule Duration to First Gas Baseline Duration to First Gas = 1,020 days Offshore Gas Production Project Entire Plan : Duration 100% % % % % % % % 2161 Hits % % % % % 2117 Cumulative Frequency 35% % % % % % % Distribution (start of interval) 0%

38 Prioritize Risks that Cause Schedule Contingency to the P-80 All Risks In Specific Risks Taken Out in Order Explain the Schedule Contingency to the P-80 P-80 Date Take Risks Out: 6-Jan-17 Days Saved % of Contingency Company's engineers may vary in experience 14-Nov % MTO, Specs may not be ready for the ITB 5-Oct % Scope growth may be more than expected 1-Jul % May have problems interfacing phases 24-Jun % HUC resources may not be experienced 19-Jun % Uncertainty Duration Estimating 20-Mar % Total Contingency % The order of risks is the best order at each step in this table. However, because of the schedule s structure some Days Saved values show inversion. 38

39 Picture of Taking Risks out in Priority Order 100% 90% Variation:91 Variation:5 Variation:7 Variation:96 Variation:40 Variation:51 80% 70% 60% 50% 40% Cumulative Probability 30% Variation:58 Variation:3 Variation:5 Variation:19 Variation:35 Variation:50 20% 10% 0% Mar May Jun Aug Oct Nov Jan Mar Apr Jun 17 39

40 Summary Cost Risk Analysis Results Cost Risk Analysis Total Project $ millions Baseline Cost 1,473 P-5 P-50 P-80 P-95 Risk Analysis Results 1,541 1,667 1,736 1,808 Dollars from Baseline Percent from Baseline 5% 13% 18% 23% 40

41 Where is the Cost Risk by Resource Cost Contingency Breakdown by Resource $ millions Resource Baseline P-80 % Contingency Project Management Team % Engineering % Installation % Fabrication % Hook Up & Commissioning % Procurement % Drilling % TOTAL PROJECT 1, % Cost Contingency Breakdown by Type of Resource $ millions Type Baseline P-80 % Contingency Materials (Equipment, raw materials) % Labor (including rented equipment) % 41

42 Risks and Inherent Uncertainty Cost Risk Analysis Results Baseline Cost $1,473 million 400 Offshore Gas Production Project Entire Plan : Cost 100% $1,945, % $1,808, % $1,776, % $1,752, With inherent uncertainty P-80 is $1,509 million Adding Risk Drivers P-80 Cost $1,736 million 18% contingency Hits % $1,735, % $1,719, % $1,707, % $1,695, % $1,685, % $1,676, % $1,666, % $1,655, % $1,645, % $1,635, % $1,624, % $1,612, Cumulative Frequency 20% $1,599, % $1,585, % $1,567, % $1,541, $1,500, $1,600, $1,700, $1,800, $1,900, Distribution (start of interval) 0% $1,449,

43 Analysis of the Sources of Cost Contingency Source of Cost Contingency at the P-80 Total Cost Contingency Total Cost All-Risks 1,736 Baseline Cost 1, Contribution to Contingency Take Out all Schedule Risks 1, Take Out all Cost Risks 1, Interaction of Cost/Schedule Risks 2 Total Contingency % of the risk to cost comes from risk to schedule, common finding 43

44 Priority Risks to Cost Measured at P-80 Prioritize Risks to Cost at the P-80 $ millions Total Project, All Risks 1,736 Baseline cost 1,473 Take Out Risk Events one at a time: $ millions saved % of total contingency Scope Growth may be more than expected 1, % Company's Engineers' may be inexperienced 1, % MTO, Specifications may not be ready ITB 1, % Quality engineers may be FAB, Suppliers 1, % Problems interfacing Phases 1, % LLE Suppliers may be busy 1, % Experienced HUC resources availability 1, % Subsea conditions may not be as expected 1, % Take out the 3-point estimates Schedule duration estimate are uncertain 1, % % 44

45 Graphical Effect of Taking the Risks out One at a Time in Priority Order 100% 90% Variation:$36, Variation:$1, Variation:$2, Variation:$5, Variation:$6, Variation:$24, Variation:$73, Variation:$54, Variation:$57, % 70% 60% 50% 40% Cumulative Probability 30% Variation:$17, Variation:$1, Variation:$ Variation:$1, Variation:$3, Variation:$9, Variation:$19, Variation:$33, Variation:$39, % 10% $1,500, $1,550, $1,600, $1,650, $1,700, $1,750, $1,800, $1,850, $1,900, % 45

46 In this case study Cost and Schedule are Correlated 51% Offshore Gas Production Project Deterministic Point Inside both limits Outside both limits 11% 7% 82% $1,900, $1,850, $1,800, Entire Plan: Cost $1,750, $1,741, $1,700, $1,650, % $1,600, PLAN $1,550, $1,500, % $1,450, Jan Apr Jun Aug Sep Nov Jan Feb Apr Jun 17 Entire Plan: Finish 11% 70% of the risk is in lower-left quadrant 46

47 Probabilistic Cash Flow Resource Flow for Cost Filter: Entire Plan Scheduled P80 2,000,000 1,800,000 1,600,000 terministic Cost: $1,473, Deterministic Finish: 20 Mar 16 1,400,000 1,200,000 1,000, , , ,000 Cumulative Cumulative Monthly Cash Flow - P-80 - Planned 200, Jan Apr Aug Nov Feb May Sep Dec Mar Jul Oct Jan Apr Aug Nov Feb Jun 17 Time 0 47

48 Analysis of a Risk Mitigation Scenario Risk Mitigation Scenario Schedule Impact Cost Impact First Probabi Most Most Gas Low High Low High lity Likely Likely Date Risk to be Mitigated Company's Engineers may be inexperienced Proposed Mitigation: Increase Salaries for company Engineers to competitive levels Company's Engineers may be inexperienced Spending $35 million for Engineers salary is assessed to reduce the probability of this risk from 85% to 20%. Because the schedule slippage is 53 days less than before, there is $21 million we do not need to reserve and so the net cost impact of the mitigation at the P-80 is offset by $21 million, for a net cost of $14 million. Before Mitigation 85% 90% 110% 130% 95% 100% 110% After Mitigation 6-Jan % 90% 100% 110% Nov- 16 Project Cost ($ million) Improvement Cost of proposed Mitigation 35 Net Improvement from Mitigation ,735 1,714 48

49 NASA Joint Confidence Level (JCL) defined For implementation of each major program segment, all space flight and information technology programs are to be baselined or rebaselined and budgeted in accordance with the following: a) There is a 70 percent probability (or a different probability that is approved by the decision authority) of achieving the stated life cycle cost and launch schedule b) Projects are to be baselined or rebaselined and budgeted at confidence level consistent with the program s confidence level 49

50 JCL is determined by the joint distribution of cost and schedule Integrated cost-schedule simulation is a perfect tool since it provides a scatter diagram of time and cost pairs that are consistent scenarios given the plan and the risks Find points where time and cost are jointly satisfied at P-70 50

51 JCL Results are on Indicated Contour Line 79% $58,000,000 9% 12% $56,000,000 $54,000,000 Joint time-cost probability = 70% $52,000,000 $50,000,000 Entire Plan: Cost $48,000,000 $46,000,000 $44,000,000 $42,000,000 $46,103,883 79% $40,000,000 $38,000,000 $36,000,000 $34,000,000 $32,000,000 70% 04 Jan 13 9% 02 Aug Sep Nov Dec Feb Apr May 13 Entire Plan: Finish Time Cost correlation is 69% 51

52 THANK YOU! 52

53 Integrated Cost Schedule Risk Analysis Using the Risk Driver Approach Qatar PMI Meeting February 19, 2014 David T. Hulett, Ph.D. Hulett & Associates, LLC 53