Part 62: Data Analysis
(Project Integration Management: Monitor and Control Project Work)
(Project Integration Management: Perform Integrated Change Control)
(Project Integration Management: Close Project or Phase)
(Project Scope Management: Plan Scope Management)
(Project Scope Management: Collect Requirements)
(Project Scope Management: Define Scope)
(Project Scope Management: Control Scope)
(Project Schedule Management: Plan Schedule Management)
(Project Schedule Management: Estimate Activity Durations)
(Project Schedule Management: Develop Schedule)
(Project Schedule Management: Control Schedule)
(Project Cost Management: Plan Cost Management)
(Project Cost Management: Estimate Costs)
(Project Cost Management: Determine Budget)
(Project Cost Management: Control Costs)
(Project Quality Management: Plan Quality Management)
(Project Quality Management: Manage Quality)
(Project Quality Management: Control Quality)
(Project Resource Management: Estimate Activity Resources)
(Project Resource Management: Acquire Resources)
(Project Resource Management: Control Resources)
(Project Risk Management: Plan Risk Management)
(Project Risk Management: Identify Risks)
(Project Risk Management: Perform Qualitative Risk Analysis)
(Project Risk Management: Perform Quantitative Risk Analysis)
(Project Risk Management: Plan Risk Responses)
(Project Risk Management: Monitor Risks)
(Project Procurement Management: Plan Procurement Management)
(Project Procurement Management: Conduct Procurements)
(Project Procurement Management: Control Procurements)
(Project Stakeholder Management: Identify Stakeholders)
(Project Stakeholder Management: Plan Stakeholder Engagement)
(Project Stakeholder Management: Monitor Stakeholder Engagement)
- Data Analysis techniques include
- Alternatives Analysis
- What is the best type of schedule for this project? If we use Rolling Wave Planning, how often should we update the schedule?
- What do we do when our project goes off track? We should have some corrective and preventative actions
- Should we make, buy, or lease a component or material?
- What are the best resource levels for this project?
- How do we balance resources, costs, and durations to obtain the best outcome?
- For example, we are designing an office building
- If we hire 10 engineers, they will complete the design in 10 months, at a cost of $1,000,000
- If we make the engineers work overtime, they will finish the design in 8 months, but it will cost $1,200,000 (we increase our cost but decrease our duration). The client might be happy if we complete the project ahead of schedule, but will their satisfaction be worth the additional $200,000 cost?
- If we hire 12 engineers, they will finish the design in 8 months, at a cost of $1,000,000 (we increased resources and decreased duration without affecting the cost). But there is a risk that we can’t find 12 qualified engineers, or that we won’t be able to train them all successfully.
- Assumption and Constraint Analysis
- Our project management plan had constraints and assumptions
- We use this tool to identify the validity of our assumptions and constraints
- Incomplete, unstable, or inaccurate assumptions lead to negative risk
- Ability to remove a constraint leads to positive risk
- Cost-Benefit Analysis
- When we have several options, we want to find the best one (not necessarily the cheapest, not necessarily the highest quality)
- The highest quality option might not be worth the cost
- We weigh the cost of each option against its potential benefit to see if it’s cost-effective
- When we have several options, we want to find the best one (not necessarily the cheapest, not necessarily the highest quality)
- Cost of Quality
- Quality costs money
- We need to figure out if increasing the quality of the project will be worth the additional investment
- If we reduce costs now, will it increase costs later (in manufacturing and/or maintenance)?
- We want to find the best quality cost. We reach a point where spending more money on quality brings no additional benefits.
- Prevention Cost
- Cost to prevent poor quality in the product/service
- Appraisal Cost
- Cost of catching defects in products after they are manufactured, but before they leave the factory
- Failure Cost
- Cost of repairing defective products that we have already given to the customers. Includes damage to our reputation
- Decision Tree Analysis
- A decision tree is like a flow chart
- It shows us the different choices we can make in the project, and their likely outcomes
- Each decision could have different costs and risks
- We can calculate the cost of each branch in order to find the best path
- Document Analysis
- Documents that can be analysed include Agreements, Business Plans, Process Flows, Marketing Literature, Issue Logs, Laws, and Requests for Proposals
- We can use Document Analysis to identify risks, especially when we identify uncertainty in documents
- Earned Value Analysis (EVA)
- Integrates the Scope Baseline, Cost Baseline, and Schedule Baseline
- Planned Value (PV)
- The Planned Value for an activity is the budget allocated to that specific activity
- At any specific time in the project, we can look at the Schedule and see how many activities should have been completed
- We can add up the budgets of those activities
- This gives us the Planned Value (i.e. how much value our project should have completed by now)
- At the end of the project, we have the total Planned Value (i.e. the value of all the activities). This is known as the Budget at Completion (BAC), or Performance Measurement Baseline (PMB).
- Earned Value (EV)
- At any specific time in the project, we can look at how much work (how many activities) has been completed
- We add up the total budget that was allocated to those activities
- That is the Earned Value (EV)
- We monitor EV to make sure that the project is on track
- Actual Cost (AC)
- At any specific time in the project, we can see how much money we spent
- That is the Actual Cost
- If the project is on budget and on schedule, the Actual Cost, Earned Value, and Planned Value will be equal
- If the Actual Cost is more than the Earned Value, the project is over-budget
- If the Earned Value is less than the Planned Value, the project is behind schedule
- Influence Diagram
- An influence diagram shows us the relationship between the project’s outcomes and influences
- Elements in the diagram with uncertainty can be represented by ranges
- We can evaluate the diagram through a simulation technique such as Monte Carlo
- Iteration Burndown Chart
- We use this chart to track the amount of work that is remaining in the Iteration Backlog
- We can use a trendline to predict the variance at completion
- We can use the trendline to forecast the completion date based on the amount of work remaining
- Performance Reviews
- We use a performance review to measure the actual performance against the schedule baseline, and to predict the duration for the remaining project tasks
- We use a performance review to measure the actual performance of a seller under contract. We can determine if the seller is behind schedule, over/under budget, or performing with poor quality.
- Process Analysis
- Identify areas for process improvements
- Used in Quality Management
- Reserve Analysis
- How much contingency and management reserve do we need for our project?
- Remember that we calculated the estimated duration and budget of our project. There will be some unknown issues that will pop up and delay our project. We add a Contingency Reserve, also known as a Schedule Reserve (for schedule) or Contingency Allowance (for budget)
- The Contingency Reserve is part of the Schedule Baseline and Cost Baseline
- The reserve is for known-unknowns such as rework
- The Schedule Reserve can be calculated as a percent of the estimated activity duration or it can be fixed
- The Contingency Allowance can be calculated as a percent of the estimated budget or it can be fixed
- As the project continues, we use up parts of the reserve, or we reduce it.
- For example, we estimate that our building will take 10 months to construct. We estimate that the exterior portion will take 5 months and the interior portion will take 5 months. We add a Contingency Reserve of 2 months. Therefore, our Schedule Baseline is 12 months.
- Why? Building materials might get delayed or the customer is unsatisfied with the work and we must redo part of the project. These are “known-unknowns”. We don’t know that they will happen, but they are easy to predict. There
- Let’s say that the exterior of the project is complete in 5 months, with no issues. Now we have 7 months to complete the interior (5 month estimated plus the remaining 2 months of the Contingency Reserve). We could reduce the Contingency Reserve to 1 month.
- Why do we want to reduce the Contingency Reserve? Why don’t we just hoard it until the end? Our Contingency is forcing employees and contractors to remain on the project for the full 12 months, when we only need them for 11. If we tell them ahead of time that their services probably won’t be required for the final month, the organization can plan to move them to another project early
- Let’s say that the exterior of the project requires some Italian marble, and the marble is delayed by 15 days, forcing us to complete the exterior in 5.5 months. Then we must deduct 15 days from our Contingency Reserve.
- We also have a Management Reserve
- The Management Reserve is not part of the Schedule Baseline/Cost Baseline
- The Management Reserve has already been approved for the schedule/budget
- It is for unknown-unknowns
- If we encounter a scenario where we need to use our Management Reserve to change our Schedule Baseline/Cost Baseline, we must follow the Change process
- Regression Analysis
- How do different variables relate to each other? If we can use one variable to predict another, we can predict the performance of the project.
- Risk Data Quality Assessment
- We use data to identify risks, their impact, and their probability of occurring
- When the data is of low quality, then our risk analysis is useless (garbage in = garbage out)
- We can identify the quality of our risk data in the following areas
- We can weigh the data quality characteristics to obtain a quality score
- Risk Probability and Impact Assessment
- For each risk
- What is the probability that a risk will occur?
- What is the impact of the risk?
- Consider the effect on schedule, cost, quality and/or performance
- We should ask project team members and external experts
- We can look at other areas besides Probability and Impact
- How quickly do we have to respond to the risk?
- For example, if a building collapses, we must respond immediately to rescue the occupants. If we discover weeds growing in our lawn, we could wait until next year to take them out.
- How long will the risk take until it impacts an objective?
- For example, if we’re building a car, and aluminum prices increase, how long will it take until the budget is impacted? If we have a huge stockpile of aluminum in our warehouse, we might be able to wait until prices drop before making another purchase, thus the proximity is low. If we order aluminum every week, then the proximity is high.
- How much time between the risk occurring and the impact being discovered?
- For example, a bridge could crack due to weather or poor-quality concrete, but nobody notices the crack until the bridge collapses ten years later.
- How easy is it to manage the occurrence of the risk?
- For example, we can’t control aluminum prices on the market, but we can negotiate a long-term, fixed-price contract with the supplier. Then we can reduce the occurrence of changes to the price.
- How easy is it to control the risk outcome?
- For example, if aluminum prices increase, and we can switch to another cheaper metal, such as magnesium, our controllability is high.
- How easy is it to detect the risk when it occurs?
- For example, if we perform regular X-rays on the bridge, it will be easy to detect a crack.
- To what extent Is the risk related to other risks?
- For example, if the bridge cracks, we must divert machines, equipment, and engineers to repair it. This reduces the resources available to complete the project, which could impact the schedule and budget.
- Strategic Impact
- How much effect does the risk have (positive or negative) on the organization’s goals?
- For example, if the bridge cracks, the company could face lawsuits and be considered negligent. The company would not be considered for future work with the government. Then the impact would be high.
- How significant do the stakeholders perceive the risk?
- The risk itself might have a minor impact on the budget and/or schedule, but a stakeholder (especially a powerful stakeholder) may see the risk in a different way.
- For example, if the bridge cracks, a government minister may want to create a news story or bar the company from future government contracts.
- For each risk
- Root Cause Analysis
- Determine the most fundamental reason or reasons for why a defect occurred
- If we remove the root cause, the defect won’t reoccur
- Sensitivity Analysis
- We are asking the question: Which risks have the largest impact on the project’s outcome?
- We calculate the correlation of each risk with the project’s outcomes
- We can use a tornado diagram to display our results
- We combine all the project’s risks and evaluate their total impact on the project’s objectives
- In a project, we have many different possible sets of risks, constraints, and scenarios, and each one has a different probability of occurring
- In a simulation, we calculate the different possible durations of each activity
- The simulation results in a probability distribution for achieving our target cost or schedule
- The most common type of quantitative risk simulation is a Monte Carlo analysis
- For cost risk, we use the project’s cost estimates
- For schedule risk, we use the schedule network diagram and duration
- We can combine the cost and schedule risks to develop an accurate quantitative risk analysis model
- How does Monte Carlo work?
- We have a bunch of input variables (cost estimates, durations, percent chance that a risk occurs) and we want to calculate some output variables (project cost, project end date)
- The computer selects a value for each input randomly (from a realistic range)
- The computer uses the input variables to calculate the outputs
- We perform this calculation thousands of times
- Each time we perform the calculation, the computer selects new values for the inputs
- Therefore, we end up with different values for the outputs each time
- Some values in the range appear more often than others
- We can plot this range in a histogram or S-curve to see the probability of each possible outcome
- We can also perform a criticality analysis. For each risk, we ask how often it appears on the critical pathway during the simulation. Remember that since calculations are random, a risk may or may not occur in some iterations.
- Stakeholder Analysis
- Provides us with a list of stakeholders and their positions, attitudes, expectations, and interests
- How the stakeholder is affected by the project
- For example, in a construction project, nearby residents are affected by the new building, pollution, increased traffic, etc. The residents should be satisfied that the project won’t negatively impact their neighborhood.
- The legal and moral rights of the stakeholder
- For example, in a construction project, nearby residents may have a legal right to oppose new construction in their neighborhood. The government has a legal right to issue or deny building permits. The government should be satisfied that the project will benefit the community.
- Assets that the stakeholder owns
- For example, in a construction project, the land owners may choose not to sell their land, which could cancel the project. The land owners should be satisfied that they are getting a good deal.
- Knowledge that the stakeholder can provide the project
- For example, in a construction project, an engineering firm could provide special knowledge to assist the construction.
- What the stakeholder provides to support the project
- Could be financial resources, human resources, or political support
- For example, in a construction project, the bank would provide a loan to finance the project. The labor union would provide workers. The bank should be satisfied that the project will be profitable and that the loan will be paid back. The union should be satisfied that worker rights will be protected.
- SWOT Analysis
- SWOT means Strengths, Weaknesses, Opportunities, and Threats
- We use SWOT to obtain more details about risks that we identified
- We need to ask the following questions
- What are the strengths and weaknesses of our organization?
- What opportunities do our strengths bring us?
- What threats do our weaknesses expose us to?
- How can we use our strengths to reduce our threats?
- Technical Performance Analysis
- We use this technique for risk management
- We create a “schedule of technical achievement”, which is a list of quantifiable measures of performance
- The measures of performance could include number of defects, number of customer complaints, brightness of a lightbulb, etc.
- We compare the actual achievement against the measures to determine the impact of risks (threats or opportunities)
- Trend Analysis
- We analyse project data for patterns
- We are looking ahead to see if the project is beginning to deviate from the schedule or budget
- Can be used to predict problems ahead of time, so that corrective action can be taken before it is too late.
- Some of the Trend Analysis tools we can use include
- We can plot our Planned Value, Earned Value, and Actual Cost
- Data can be plotted weekly, monthly, or yearly
- A chart helps us visualize our data to determine how we are performing (budget and schedule)
- After we collect enough project performance data (cost and schedule), we can calculate forecasts
- The Estimate at Completion (EAC) is what we forecast the project will cost. The Budget at Completion (BAC) is what we originally forecast the project will cost.
- The EAC is the total actual costs incurred plus the estimated cost of completing (ETC) the remaining work
- We calculate the ETC with a bottom-up approach (add up the estimated cost of each of the remaining project activities)
- EAC = AC + Bottom-Up ETC
- We continue to update forecasts based on new project data
- There are three ways to calculate EAC
- Assume that the remainder of the project will be completed at the budgeted rate
- We accept the Actual Costs for portion of the project already completed
- We add the budgeted cost of the uncompleted activities to the Actual Costs
- If the activities already completed are not on budget, we are assuming that the remainder of the project will be on budget
- The remainder of the project’s cost is the Budget at Completion, less the Earned Value
- EAC = AC + (BAC – EV)
- Assume that the remainder of the project will be completed at the current CPI
- We assume that the project’s budget performance will continue at the same rate
- EAC = BAC / CPI
- For example, If the project is 10% over budget now, it will finish at 10% over budget
- Assume that the remainder of the project will be completed at the current SPI and CPI
- We assume that the project’s budget performance and schedule performance will continue at the same rate
- If a project is behind schedule, its poor schedule performance can also affect its budget
- EAC = AC + (BAC – EV) / (SPI + CPI)
- Assume that the remainder of the project will be completed at the budgeted rate
- Variance Analysis
- The Variance is the difference between the planned and actual performance
- Variance Analysis explains why the project is not on budget and/or on schedule
- It includes the cause (why?), impact (what is affected?), and corrective actions
- Can include duration, cost, resource use, and quality
- In Monitor and Control Project Work, we look at all the project’s variances to see what preventative and corrective actions can be taken
- We compare actual start and finish dates with planned start and finish dates
- We compare actual durations with planned durations
- We use Variance Analysis to figure out how much variance there is and what caused it
- We can decide how the variances will affect future work, and decide what kind of corrective action is required
- The Cost Variance is the difference between the Earned Value and the Actual Cost
- CV = EV – AC
- Schedule Variance tells us if the project is on budget, overbudget, or underbudget
- For example, if our Earned Value is $1000, and our Actual Cost is $2000, then our Cost Variance is $1000. We completed $1000 worth of work, but it cost us $2000, so we are $1000 over budget.
- The Schedule Variance is the difference between the Earned Value and Planned Value
- SV = EV – PV
- Schedule Variance tells us if the schedule is ahead of schedule, behind schedule, or on time.
- When the project is complete, the Schedule Variance is 0. Why? At the end of the project the entire project is complete. It’s not necessarily completed on schedule or on budget.
- For example, our Earned Value is $1000, and our Planned Value is $2000, then our Schedule Variance is $1000. We planned to complete $2000 worth of work, but we only completed $1000. So we are “$1000” worth of work behind schedule.
- The Schedule Performance Index (SPI) is a helpful number
- Ratio of Earned Value to Planned Value
- SPI = EV / PV
- If the project is on schedule, the SPI should equal one. That means we completed exactly the amount of work that we planned to.
- If SPI is less than one, then we are behind schedule. If SPI is greater than one, we are ahead of schedule.
- We should check our critical pathway as well (see the example below)
- Remember that we can have multiple pathways in our project’s Schedule Network, including the critical pathway
- Let’s say that we have a project worth $1,000,000, with a schedule of one year.
- Let’s say that the activities in the Critical Pathway are 10% of the project ($100,000) and the other activities are 90% ($900,000) of the project
- The Critical Pathway activities are expected to take eight months to complete
- Halfway through the schedule (six months), we complete the 90% of activities outside of the Critical Pathway. At this point, we only expected to complete $500,000 worth of activities
- Our SPI = $900,000 / $500,000 = 1.8
- Even though our SPI is 1.8, we are still two months behind schedule. Why? Our Critical Pathway activities should take eight months to complete. We haven’t started on them yet, and we only have six months left.
- So we should also check the Critical Pathway when calculating SPI
- The Cost Performance Index (CPI) is a helpful number
- Ratio of Earned Value to Actual Cost
- CPI = EV / AC
- If the project is on budget, the CPI should equal one. That means we completed exactly the amount of work that we planned to.
- If CPI is less than one, then we are underbudget. If CPI is greater than one, we are over budget.
- What-if Scenario
- We gather several scenarios and predict their effect on the project
- We can use the Schedule Network Analysis technique to calculate the effect on the schedule for each scenario
- Scenarios can include a labor strike, a missing component, or a delay in obtaining a license/permit from the government
- This data is used to decide on possible responses
- Alternatives Analysis
Consider this example
Your project is to construct an office building with 5 floors.
Each floor costs $100, for a total budget of $500.
The project will be completed in five days. Assume that we will build one floor each day.
At the end of the second day, one floor has been constructed and $400 has been spent.
What is the Planned Value?
The Planned Value is $200.
At the end of the second day, we planned to have built two floors, for a total value of $200.
What is the Earned Value?
The Earned Value is $100.
At the end of the second day, we only built one floor. Each floor is worth $100, so we only have $100 in earned value.
What is the Actual Cost?
The Actual Cost is $400.
We spent $400, so $400 is the actual cost.
What is the Schedule Variance?
The Schedule Variance is -$100.
SV = EV – PV = $100 – $200 = -$100
We planned to complete $200 worth of work, but only completed $100, so we are behind schedule by $100 worth of work.
What is the Cost Variance?
The Cost Variance is -$300.
CV = EV – AC = $100 – $400 = $-300
We completed $100 worth of work, but spent $400.
What is the Schedule Performance Index?
The Schedule Performance Index is 0.5
SPI = EV/PV = 100/200 = 0.5
We completed 50% of the work that we planned to complete.
What is the Cost Performance Index?
The Cost Performance Index is 0.25.
CPI = EV/AC = 100/400.
We are over budget by 4 times.
What is the Estimate At Completion?
The Estimate at Completion is $2000.
EAC = BAC/CPI = $500/(0.25) = $2000
We are overbudget by 4 times. Or in other words, It appears that each floor is costing us $400 to build. To build 5 floors, it will cost $2000.
What is the To-Complete Performance Index?
The To-Complete Performance Index is 4.
EAC = (BAC-EV)/(BAC-AC) = (500 – 100)/(500 – 400) = 4
We have $100 left to complete $400 worth of work. That means we must work 4 times more economically to complete the project on budget.
What is the To-Complete Performance Index, if we are provided with a new Estimate at Completion of $1500?
The To-Complete Performance Index is 0.37.
EAC = (BAC-EV)/(EAC-AC) = (500 – 100)/(1500 – 400) = 0.37
We now have $1500 to complete $400 worth of work. So we must work 37% as economically to complete the project within the new estimate. Keep in mind, we were only working 25% as economically to get to where we were.
If we complete the project at our $1500 Estimate, what is the Variance at Completion?
The Variance at Completion is -$1000.
VAC = BAC – EAC = 500 – 1500 = -1000
We were over budget by $1000.
A Summary of the Financial Tools
|Abbreviation||Name||Definition||How Used||Equations||Interpretation of Result|
|PV||Planned Value||The authorized budget for the work||The value of the work planned to be completed at a point in time|
|EV||Earned Value||The measure of work performed expressed in terms of budget authorized for the work||Planned value of the work completed to a point in time, without reference to actual costs||EV = sum of the planned value of completed work|
|AC||Actual Cost||The actual cost incurred for the work performed during a specific time period||The actual cost of all the work completed to a point in time|
|BAC||Budget at Completion||The sum of all budgets established for the work to be performed||The value of the total planned work; the project cost baseline|
|CV||Cost Variance||The amount of budget deficit or surplus at a given point in time, calculated as the difference between the earned value and actual cost||The difference between the value of the work completed, and the actual costs||CV = EV – AC||Positive = Under planned cost Neutral = On Cost Negative = Over planned cost|
|SV||Schedule Variance||The amount by which the project is ahead or behind the planned delivery date, at a given point in time, calculated as the difference between the earned value and the planned value||The difference between the work completed at a point in time and the work planned to be completed at the same point in time||SV = EV – PV||Positive = Ahead of schedule Neutral = On Schedule Negative = Behind Schedule|
|VAC||Variance at Completion||Projection of the amount of budget deficit or surplus, calculated as the difference between the budget at completion and the estimate at completion||Estimated difference in cost at the end of the project||VAC = BAC – EAC||Positive = Under planned Cost Neutral = On planned Cost Negative = Over planned Cost|
|CPI||Cost Performance Index||Measure of the efficiency of budgeted resources; calculated as the ratio of earned value to actual cost||CPI = EV/AC||Greater than 1.0 = under planned cost Exactly 1.0 = on planned cost Less than 1.0 = over planned cost|
|SPI||Schedule Performance Index||Schedule efficiency expressed as the ratio of earned value to planned value||SPI = EV/PV||Greater than 1.0 = ahead of schedule Exactly 1.0 = On schedule Less than 1.0 = Behind schedule|
|EAC||Estimate at Completion||Expected total cost of completing all work; calculated as the cost of the actual work plus the estimate to complete||If CPI is expected to stay the same until the project is complete||EAC = BAC/CPI|
|If future work will be at the planned rate||EAC = AC + BAC – EV|
|If initial plan is no longer valid||EAC = AC + Bottom-up ETC|
|If both CPI and SPI influence the remaining work||EAC = AC + [(BAC – EV)/(CPI + SP)]|
|ETC||Estimate to Complete||The expected cost to complete the remaining project work.||Assuming work is on plan, the cost of completing work||ETC = EAC – AC|
|Reestimate the remaining work from the bottom up||ETC = Reestimate|
|TCPI||To Complete Performance Index||A measure of the cost performance that must be achieved with the remaining resources in order to meet the goal||The efficiency that must be maintained to complete the work on plan||TCPI = (BAC – EV) / (BAC – AC)||Over 1.0 = Harder to complete Exactly 1.0 = Same to complete Less than 1.0 = Easier to complete|