Formulas for Work Rates, Estimating Revised Values and the relationship to SPI

In some situations one may run accross that the initial work rate for a open price contract is going to fast or too slow so it is necessary to vary the rate after the measurement. The formulas above show the simple case where two periods P1 and P2 add up to the total period Pt.

The initial estimate to do work Wt is that a rate of Ro for the time Pt is required. If this rate is used the SPI will be 1 since the planned value will equal the earned value.

On the other hand if we have a rate R1 during P1 which is too low or too high (as in the graph) we need to reduce the rate R2 during P2 in order to reach the goal. The value of R2 to reach the goal is given in formula (5).

The SPI at any point during P1 with rate R1 is obviously higher than 1 (for the example in the graph) and is constant for the period since EV/PV=R1*t/Ro*t= R1/R0.

The SPI during the second period P2 with R2, however, is variable and varies from R1/R0 at the begining of the period to 1 at the end since we calculated R2 with just this purpose in mind. The exact formula can be found by combining the EV/PV from

and equation (5). The result leads to a variable SPI for the period which is somewhat missleading in the sense that if we are using SPI as an indicator it serves us well as we drift from the schedule during P1 and gives us a sense of how much we need to correct during that period.
But once we have started the correction, however, as we have in the second period SPI varies and shows us how close we are to getting to the schedule.
To see the overall trends and for forecasting therefore it is more interesting to use R1/Ro or R2/R0 ratios and not use SPI.

The Deming PDCA Cycle and Recommended Corrective Action in PMBok

Andy Crowe in his PMP material, discussing Scope Verification, does an excellent job of discussing the difference between Requested Changes and Recommeneded Corrective Action. The key concept being that recommended corrective actions deals with effecting future results. The important question being how do we improve the plan in order to make sure the results will be more aligned to the planned values.

This is very similar to Plan Do Check Act (PDCA) cycle of Deming. Requested changes do not move the activity in the PDCA cycle, that is to say the changes are cimply carried out according to plan while recommened corrective action is an input to the plan and hence moves the data through the PDCA cycle.

A few Sample PMP Questions

Here are a few sample PMPs I made to help remember a few interesting issues.

1. You are the PM of a 3.5 million dollar Project to develop a new brake system for an automotive parts supplier. You have been on vacation for a week and arrive back at work to find a stack of papers of your desk with requested changes, work performance information, recommended corrective actions and deliverables. Your boss comes in to your office with a nervous look on his face and says, “I need those validated defect repair reports ASAP.” What process should you start working on:

A) Manage Stakeholders
B) Integrated change control
C) Product Analysis
D) Constrained optimization

2. You are managing a small project to create a manual for a new mobile telephone. You are reviewing the contract documentation and the contract management plan to produce the closed contract. Which of the following tools and techniques might be a good idea to use?

A) Strong Matrix
B) Constrained Optimization
C) Procurement audits
D) Updates to the contract file as part of the Organizational Process Assets

3. You are managing a large integration project for the EU commission including 9 partners from 5 European countries. You are using the project scope statement and the project scope management plan and are decomposing the project deliverables into work packages with the help of templates. Which process are you carrying out?

A) Fast Tracking
B) Scope Control
C) Activity Definition
D) Qualitative Activity Estimation

4. You have just carried out activity definition and are using the milestone list and activity list to create the project schedule network diagram. A team member suggests that you use the precedence diagramming method. You show him your project management scheduling software which does not support “dummies” and he suggests an alternative to this tool and technique called:

A) Activity on node diagram
B) Activity on arrow diagram
C) Discretionary dependency analysis
D) Published Estimating data

5. You are carrying out the activity resource estimating process. Which of the following is not a tool and technique for this process?

A) Expert Judgement
B) Resource Breakdown Structure
C) Published Estimating data
D) Bottom-up Estimating

6. Which of the following processes does not use the Project Management Plan or its subsidiary plans as an input(s)?

A) Scope Definition
B) Integrated change control
C) Direct and Manage Project Execution
D) Activity Sequencing

7. Which of the following pair of project management plan components are created in the develop Project Management Plan process?

A) Quality baseline and Cost baseline

B) Cost Management plan and Communications management plan

C) Cost Management plan and Schedule management plan

D) Scope Management Plan and Schedule management plan

Answers:
1B2C3C4B5B6D7C

Project Management Methodology as a Tool and Technique

Lets get the discussion started with a PMP exam Sample Question:

You are managing a project to develop a preliminary risk plan for another project. You are using a project management methodology and a project management information system to create an implemented change request. What process are you performing?

A) Scope definition
B) Initial risk scope evaluation
C) Expert judgement
D) Direct and Manage Project Execution

OK So that is pretty easy (the answer is at the end of the entry) but what I really noticing when I made this question was the following:

The PMI considers a project management methodology as a tool and technique and Software development frameworks as inputs as part of the organizational process assets.

You need both but since we are working with a project management standard (the PMBok is a standard) the project management methodology is a tool and technique but the software development process framework is an organizational process asset because it is selected specifically for the the organization and industry The link is weaker. Tools and techniques are related to project management not a specific industry.

But how do all these things fit together. Above I have included an image which fits the pieces together. For software the product lifecycle includes operations processes and even teh project lifecycle overlaps with some operations processes as part of the deployment.

The answer to the sample question is D, you are in Direct and Manage Project Execution since that is the only place where you implement changes.

A more complex question is: If I have a project management plan why do I need a project management methodology as a tool and technique? The reasonable answer is probably that not everything will be detailed in the project management plan and hence one will sometimes have to refer to the methodology. I´m not sure that is too convincing though.

Anti-buffering: Moving risk from schedule activities to the contingency reserves

Anti-buffering the project is an excellent way to motivate risk management. Many PMs are unimpressed by risk management because they don't see the value of putting risks in an spreadsheet that they look at a few times in the project. While even this can give some possible results it doesnt go far to generate real value.


Completing the PMBok risk processes is another story all together - following these processes generate much more value. One of the main values are the relationships between schedule development and risk analysis through the contingency reserve. A main result is that the schedule contingency reserve is where we should put the risk of specific activites.


Lets take an example in a project I am managing:

I have a task which is to implement a PBX functionality called "whisper mode" for a call center.


When I estimated this task intially I put in some buffer for risk since I had only recieved one high level estimate. You could say that at this point my risks where unknown.


As I started to investigate the design and possibilities the risks become clearer. One design called for more software and was more complicated. The other solution uses built in functionality. So lets put that this way:

• Solution A: Cost 10, Time 2 - probablity that I can use this one 30%.


• Solution B: Cost 20, Time 20 - probablity that I can use this 70%.

So how do I include this in schedule development and risk processes?


I put the minimum time in the schedule and then put the difference in the contingency buffer. In other words I am saying that there is a risk I may have to use solution B.





So I will put in my risk register "Risk: Have to use Solution B" with a schedule impact of 18 and a cost impact of 10 and a probablity of 70%.

This effectively removes the buffering in the activity and moves it to the contigency buffers.



And I will put in my contingency schedule buffer 18 days and my contingency budget buffer 10 and leave my activity at time 2 with a cost of 10.

Now somebody might say this is very missleading. Wouldnt it be better to put the more probable result in the schedule and then put a opportunity in the risk register. In other words put the time cost at 20 and time at 20 and then put an oppertunity of 18 days and 10 cost units in the contingency reserves.

Hmm...dont think so. One cant have negative contigency reserves and the effect is to buffer the schedule activity to a maximum which as we know from Critical Chain is not a good idea.

Calculating Probabilities, Risk, Decision Trees and Monetary Value

There are several areas where you might need to calculate probablities of an occurance. Most likely they are involved with calculating risk.

One of those areas is calculating the advantage of splitting an order. If you have an activity on the critical path and in order to start working you need at least part of the order completed it might be a good idea to split the order between two providers. While most of us think of this in the manufacturing scenario it is very applicable to software development. The calculation can be thought of in two ways:

• If provider A is on time 80% of the time and company B is on time 70% of the time then the overall assurance that I will get at least something is 80% + (20% x 70%) = 94%.



• The general parellel calcualtion rules is that for two components in parallel the availability is A = 1-(1-Ac)^2 where Ac is the availability of each component and they are equal.



• The more general case of this formula is then 1-(1-Ac1)*(1-Ac2) or 1-(1-.80)*(1-.70)=94%

A good question is how does this relate to decision trees. In a decision tree there are normally two types of branches/nodes the decision node and the chance node. The decision node is mutually exclusive. Mutually exclusive means that it is either A or B not A nor B or A and B. In other words the probablity of A and B should add up to 1. The results on the branch are usually done with expected monetary value where the probablity can add up to more than 1 or less than one since these are not exculsive and are simply possiblities which generate value on the node.

Can we have a parellel node on the chance node of a decision tree? Yes absolutely. But then we dont use the simple expected monetary value instead we use the 1- (1-Ac1)*(1-Ac2) formula above to find the monetary value.