A Practical Way To Learn About EVM ***The Deck Project***
Back to the Baseline 2
The Deck Project The product of this project is a new second-story deck in the backyard. This new deck will replace an existing deck, stand about 10 feet off the ground, and measure 12 feet by 12 feet square. On one side, a set of stairs will lead to the ground. The structure will be attached to the house and supported with four posts resting on concrete footings. Due to city ordinances, the footings must extend below 4 feet. The homeowner and friends (when available) will construct the deck with standard materials. All supplies will be ordered from and delivered by the local home improvement center. 3
Schedule Baseline #1 Activity Planned Value Planned Scheduled Finish (Effort-Hours) Duration (Days) (Day) Draw plans and material lists 6 1 Day 1 Destroy old deck 12 1 Day 2 Truck off trash 4 0.5 Day 3 Purchase materials 4 0.5 Day 3 Dig four footing holes 8 2 Day 5 Pour concrete footings 4 1 Day 6 Construct main deck 12 3 Day 9 Constuct railings and stairs 8 2 Day 11 Clean-up 6 1 Day 12 Actual Costs (Effort-Hours) Status (Done/Started) Earned Value (Effort-Hours) Totals: 64 12 0 0 0 0 4
The Project Schedule 5
For our deck project, the total duration is 12 days with 64 hours of total effort. Figure 2 shows the effort hours by day (as bars) and the cumulative effort hours (as a line). The height of the line on Day 12 is 64 hours, which corresponds to our total cost. 6
It s Day 5, So What s The Problem? # Activity Planned Value (Effort-Hours) Planned Duration (Days) Scheduled Finish (Day) Actual Costs (Effort-Hours) Status (Done/Started) Earned Value (Effort-Hours) 1 Draw plans and material lists 6 1 Day 1 6 Done 6 2 Destroy old deck 12 1 Day 2 12 Done 12 3 Truck off trash 4 0.5 Day 3 4 Done 4 4 Purchase materials 4 0.5 Day 3 4 Done 4 5 Dig four footing holes 8 2 Day 5 10 Started 0 Total-To-Date 34 5 36 26 6 Pour concrete footings 4 1 Day 6 7 Construct main deck 12 3 Day 9 8 Constuct railings and stairs 8 2 Day 11 9 Clean-up 6 1 Day 12 Totals: 64 12 0 36 0 26 7
Summary According to the schedule, we had planned to expend 34 hours of effort on the project (the sum of the planned cost for the first five activities). Those 34 hours were to complete the first five activities. However, based on actual results, we have expended 36 hours and completed the first four activities. Activity 5, Dig footings, has started, but is not yet finished and it is running over plan (the planned effort was 8 hours, recorded actuals show 10 hours so far). And since the digging is not done, we can t take any credit for Earned Value on the activity. 8
So what s the status? At the end of Day 5, we planned to accomplish 34 hours of work We ve expended 36 hours, yet we ve only earned 26 hours Planned Value (Cumulative) = 34 Hours Actual Costs (Cumulative) = 36 Hours Earned Value (Cumulative) = 26 Hours 9
Calculate CPI & SPI Group Exercise How would you explain these indexes from the perspective of this project What would the graph of this project look like? Draw a high-level graph What does this graph tell us? 10
Calculating EVA CPI = 26/36 = 0.72 SPI = 26/34 = 0.76 Based on our CPI, we are realizing 7.2 hours of result for every 10 hours of actual effort expended on our project. Based on our SPI, we are realizing 7.6 hours of result for every 10 hours of planned effort on our project. 11
2.00 Cost Performance Index (CPI) 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 Day 5 0.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 Schedule Performance Performance Index (SPI) 12
Planned Value, Actual Cost & Earned Value vs. Project Duration Through Day 5 13
What the graph tell us By comparing the Actual Cost line to the Planned Value line, we can see that, through Day 3 of the project, actual costs were according to plan. But on Day 4, the red jumps above the blue, showing that we ve gone over plan. Also, because no work was marked Done on Days 4 or 5, the Earned Value (green line) levels off. The gap between Actual Cost, Planned Value, and Earned Value clearly indicates project problems. 14
Group/Homework Exercise What is the EAC of this project? Hint: What s the assumption of the CPI at this point? Can your figure out what the new forecasted end date will be? Hint: What s the Duration at Completion [DAC]? The PMBOK forgets that EAC calculates the vertical axis, but needs an additional EAC[t] (t = time ) calculation for the horizontal axis What would the graph look like? 15
EAC[t]: Estimated Time At Completion In order to calculate the EAC[t], we need to add additional variables to our EVM collection: Planned Duration (PD) - The planned duration at a given status period Actual Duration (AD) The actual duration expended Earned Duration (ED) The duration earned Duration At Complete (DAC) The sum of all authorized durations allocated to the project 16
EAC[t] Details The EAC[t] equals the actual durations incurred, plus the estimated durations for completing all the remaining project work. The use of a particular EAC[t] should be exactly the same as the conditions for the EAC, with the exception of the assumption of the SPI, rather than the CPI, remaining constant throughout the project. Outlined in the Practice Standard for Earned Value Management published by PMI in 2005, the SPI and using what they refer to as the average PV per unit of time can be used to provide a rough estimate of when the project will complete or if the project s performance trend stays fairly constant, what the forecast end date could be. 17
EAC[t] Details Cont d The equation is as follows (*Note: PMI uses PD for planned duration instead of DAC. I feel that DAC is more accurate counterpart to BAC): This can equation can be simplified to: Therefore, the most likely EAC[t] is DAC SPI which is the direct counter part to formula #3 of EAC which is BAC CPI. 18
Where are we headed? If these ratios are accurate, we can use them to forecast completion of the project. To forecast the estimate at completion (EAC), divide the original cost by the CPI, with the assumption that the variance should stay the same. For our project, EAC = 64 / 0.72 = 88.6 or 89 hours. The estimate at completion for time (EAC[t]) is the original duration divided by the SPI. Again, for our project, the EAC[t] = DAC/SPI = 12 / 0.76 = 15.7 or 16 days (see next graph). 19
Forecast of Project EAC & EAC[t] 20
What s the implications of the data? What s happening? So after five days of work on the project, it looks like we ve added about 4 days to the schedule and 25 hours of work to the budget. Is the project really that far off track? (CPI and SPI are conservative forecasting tools each assumes that past performance is an indicator of future results.) After a review and discussion of the project s performance, our key stakeholders discovered that the original estimates for digging were based on favorable soil conditions. For three of the four holes, this was a valid assumption. However, the diggers uncovered a large boulder (more than 500 lbs) on the fourth hole. 21
Group Exercise Why did we only choose the EAC calculation of EAC = BAC/CPI (64/0.72)? What other EAC calculations could we have used and why? What would the graph look like if we plotted all of the scenarios and how would you present this? Hint: think of the three-option rule (Arco gas demo) What other forecasting equation would be useful to use and why? Hint: What about TCPI? What would the graph for TCPI look like? 22
Additional Analysis The PMBOK lists four types of EAC calculations The formula EAC = AC + Bottom-up ETC is the most accurate, but would require you to manually rebuild the work package estimates from the ground up. Who has that time! We therefore have three other EAC equations we can utilize, but why would we use them and not the one we already calculated? 23
Statistical EAC Comparisons Low-end over-run to date The optimistic or best case EAC = AC + BAC EV 36 + 64 26 = 74 Middle range EAC The most likely case EAC = BAC/CPI 64/0.72 = 89 High-end range EAC The pessimistic or worst case EAC = AC + [(BAC EV)/(CPI x SPI)] 36 + ((64 26)/(0.72 x 0.76)) = 105 24
Statistical EAC[t] Comparisons (Time) Low-end over-run to date The optimistic or best case EAC[t] = AD + DAC ED 5 + 12 4 = 13 Middle range EAC The most likely case EAC[t] = DAC/SPI 12/0.76 = 15.7 High-end range EAC The pessimistic or worst case EAC[t] = AD + [(DAC ED)/(CPI x SPI)] 5 + ((12 4)/(0.72 x 0.76)) = 20 25
Graph with three point EACs EAC = AC + [(BAC EV)/(CPI x SPI)] EAC = BAC/CPI EAC = AC + BAC EV 26
27
What about TCPI? The To Complete Performance Index (TCPI) provides the performance the CPI must maintain in order to get back to 1.0 at a particular status date At our current status day, we are more concerned with the question of is the BAC achievable? rather than is the current EAC realistic and achievable? So we use TCPI[p] = (BAC EV)/(BAC AC) or (64 26)/(64 36) = 1.36 In other words, given where we are at, we would need to maintain a CPI of 1.36 to get back to 1.0 28
Graph of TCPI[p] Sunk Costs Opportunity Costs + 1.36 TCPI[p] 1.00 Baseline - CPI 0.72 Day 5 Day 12 29
Graph of TCPI[p] & TCPI[e] Sunk Costs Opportunity Costs + 1.00-1.36 1.0 0.72 0.55 0.72 (CPI) TCPI[p] otcpi[e] mtcpi[e] ptcpi[e] Day 5 Day 12 30
What about TCPI[t]? Just as there is a counterpart to EAC with EAC[t] for time, we can compute the TCPI[t] for the horizontal axis: TCPI[t][p] = (DAC-ED) (DAC-AD) TCPI[t][e] = (DAC-ED) (EAC[t]-AD) At our current status day, we are more concerned with the question of is the DAC achievable? rather than is the current EAC[t] realistic and achievable? So we use TCPI[t][p] = (DAC ED)/(DAC AD) or (12 4)/(12 5) = 1.14 In other words, given where we are at, we would need to maintain a SPI of 1.14 to get back to 1.0 31
Graph of TCPI[t][p] Sunk Time Opportunity Time + 1.14 TCPI[t][p] 1.00 - SPI 0.76 Day 5 Day 12 Baseline 32
Graph of TCPI[t][p] & TCPI[t][e] Sunk Time Opportunity Time + 1.00-1.14 1.0 0.72 0.53 TCPI[t][p] otcpi[t][e] mtcpi[t][e] ptcpi[t][e] 0.76 (SPI) Day 5 Day 12 33
Group Exercise What corrective actions would you take to get the project back on track? Use the grid on the following slide to outline the possible corrective actions you would take based on product features and performance (scope), schedule and cost Hints: Re-evaluate how you would re-scope the digging of the 4 holes to get the project back on track How does this effect the CPI & SPI Does this have any effect on the EAC and the overall duration of the project? 34
Possible Corrective Action for the Deck Project 35
Possible Corrective Action for the Deck Project 36
Strategies to dig out way out So our project is not quite as badly off as the numbers suggest. Three of the four holes are complete, and the fourth hole would be done if not for the boulder. If we informally apply a % complete evaluation to our project, we earn 75% of the hole digging activity. This increases our Earned Value to 32 hours, and our CPI and SPI to 0.89 and 0.94 respectively. If we re-compute the EAC and schedule, our EAC forecast improves from 88 to 72 hours and we shorten our schedule estimate from 15.7 days to 12.7 days. This information will help us decide on what corrective action to take. 37
Re-baseline the project! After some discussion, the stakeholders decided that the key constraint is the product s features and performance. They chose to remove the bolder and place the footing where originally planned, satisfying all building codes. However, since this choice adds work to the project, the plan must be adjusted. Removing the rock involved manually excavating around it, dragging it out of the hole, and restoring the earth displaced by the rock. In total, an additional eight hours of effort and one day of schedule need to be added. 38
# Activity Planned Value (Effort-Hours) Planned Duration (Days) Scheduled Finish (Day) Actual Costs (Effort-Hours) Status (Done/Started) Earned Value (Effort-Hours) 1 Draw plans and material lists 6 1 Day 1 6 Done 6 2 Destroy old deck 12 1 Day 2 12 Done 12 3 Truck off trash 4 0.5 Day 3 4 Done 4 4 Purchase materials 4 0.5 Day 3 4 Done 4 5 Dig three footing holes 8 2 Day 5 10 Started 8 6 Remove bolder and dig footing hole 8 1 Day 6 7 Pour concrete footings 4 1 Day 7 (6) 8 Construct main deck 12 3 Day 10 (9) 9 Constuct railings and stairs 8 2 Day 12 (11) 10 Clean-up 6 1 Day 12 (13) Totals: 72 13 0 36 0 34 Re-baselining the project and updating the plan reflects the added task Remove boulder. No other project activities have been changed. Also, the activity Dig footings was marked Done and the value of that work counted as Earned. 39
Group Exercise What would be the new baseline results? What would the new graph look like? 40
New baseline results The effect of the re-baseline can be evaluated quickly with CPI, SPI, and their related projections. At the end of Day 5, we have a new CPI = 34/36 = 0.94 and the new SPI = 34/34 = 1.0. The new EAC = 72/0.94 = 77 hours. The projected duration is the same as the re-baselined plan, or 13/1.0 = 13. The project is still forecasted to exceed the new budget (because CPI is less than 1.0), due to two hours of rock removal on Day 5 (two hours against the eight hours added during the re-baseline). We expect that after Day 6, CPI productivity will climb back to 1.0. Re-baselining is only legitimate when it is performed in response to a stakeholder- approved scope change. 41
Original Baseline and Re-Baselined Planned Value 42
2.00 Cost Performance Index (CPI) 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 Day 5 Day 5 (New) 0.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 Schedule Performance Performance Index (SPI) 43
44
Probabilistic EVM Modeling! 45
Questions? 46