website paiken:research

Project Decision Analysis Context: Decision

by Bill Girling

According to the initial assessments, the project decision is a "GO".

On the PROJECT DECISION TREE, the two high value weights, "Legislative Support" and "Within Budget" are of particular interest because the weight assignments are so unstable. Alternate sources of funding and multi-year budget cycles can create a very "fluid" budget even when "Legislative Support" remains constant. To assist in the assessment of potential changes to the budget, the CDF generated for the "High Risk Analysis", Exhibit #6b, can be used. For example, there is a 81% probability that the combined cost for consulting and software services will be less than $5 million. To further test these relationships, a "Sensitivity Test" area was constructed on the decision tree that accommodated changes between elements, i.e. value (weight) of the economy verses value of the "majority political party" and across the decision, "Legislative Support" verses "No Support". The current assessment for "Legislative Support" assigns a value of 100 to "Go" and 25 to "No GO". It should be noted that the proportional relationship to the other attributes, "Budget" and "Deployment" are the same.

However, consider a real life example. Changes have occurred that reflect information, that was received after the initial assessments. For the attribute of "Scandal", which is an element of the "No Legislative Support" branch of the uncertain event "Legislative Support, a probability of 20% and weight of 5 was assigned. This assessment was based on the frequency and impact of "scandals" over the previous 16 years. Recently, two new "scandalous" events (Exhibit 8, and Exhibit 9) erupted, in addition to Department of Social Services "scandal" that occurred less than two years ago. In one of the situations, on April 24, 1997, the Department Medical Assistance Services which administers the highly volatile, Medicaid program made headlines by canceling a $45 million automated system acquisition effort. The Commonwealth is threatening to sue an internationally recognized vendor to "recover damages". Key government officials unconnected to these events, have already met to discuss possible impact on the project. With three such incidents within the past four years it would be very easy to justify changing the "Scandal" probability to 75%. The "weight" is becoming more of a factor than had been assessed earlier and is starting an upward climb as well. This consideration would probably be different if we were not six months away from a gubernatorial election. During the four year administrative term, sensitivity to a "Scandal" builds to an all time high during the period preceding an election. If the "weight" moves to 150 points, the decision for the project will shift to a "No GO" status, since legislative support for the next budget will at best be delayed. The is an opportunity to exercise similar scenarios in the "Budget" area as well. This is particularly valuable as the project proceeds and cost estimates become more accurate. Such "perfect information" can be considered in conjunction with the other attributes that define this decision.

---

Navigate from here: returning to the project project home page, or jump to another part of the decision analysis case study (Situation Description, Possible Decision Outcomes, Decision Analysis, Constructing & Linking The Uncertain Events, Decision, Summary) - or jump to Peter Aiken's home page, download some reengineering articles, or access other reengineering links.