Project: Project Manager

An consideration of possible courses for Eric Silver to become an optimal project manager with time and monetary constraints

Completed and performed by Eric Silver, 12/1/1999

Chapter	Purpose
1

Intent:

I will demonstrate my ability to use a diverse set of software tools, paired with the concepts of project management, to outline a course of action for developing my own Project Management skills. A living version of this endeavor should be available online at http://www.silverhouse.net/plan/index.htm at the time that this report is delivered, although the online documentation is focused on a disparate course of action.

Scope:

Because of time constraints, this report is limited in scope. 10 third level values were compared against 15 possible courses of action with rough estimates as to their expected cost and duration. Additionally, since these tasks will tend to overlap in time, I have decided to make the unreasonable assumption that there will be negligible scheduling constraints.

Shortcomings:

By deciding to use the demo version of Criterium Decision Plus I limited myself to 20 objects. Instead of creating separate files for each sub-goal and extensively considering the options available to me, I instead used the SMART method to determine relative values for each of only 5 alternatives at a time, and then used AHP to ‘norm’ these results. This process was somewhat sloppy, and because I don’t understand the SMART methodology, intrinsically imperfect.

Additionally, since the SMART method doesn’t have the ability to judge the certainty of your inputs, I am missing an additional ‘check’ to my inputs. Likely, the output is extremely skewed by imprecise inputs.

I have also scheduled myself as being 100% available for the next three years, which may be an unreasonable assumption as I have not accounted for illness or personal requirements.

Chapter	Methodology
2

Creating requirements hierarchy:

Beginning with the material available within the Project Management and Project Leadership binder, a needs hierarchy was constructed. The size of this hierarchy was constrained by my choice to use Criterium Decision Plus to 20 objects, which I pared down to 15 so that I could use it to evaluate possible decisions. The resulting hierarchy with 5 different possible tasks is shown below:

Assigning Priority values:

By fusing together indicators that I have received during the Project Management programs at UCSC, I

Goal Level	Weights	Level 2	Priorities	Rating Set	Weighted Value
Goal	0.15	Business Skills	0.25	Legal Skills	0.0375
			0.75	Organizational Skills	0.1125
	0.5	Leadership Skills	0.333	Creating Vision	0.1665
			0.333	Influencing People	0.1665
			0.333	Teambuilding Skills	0.1665
	0.25	Decision Making Skills	0.4	Planning Skills	0.1
			0.6	Requirements Analysis Skills	0.15
	0.15	Tracking Skills	0.25	Software tool familiarity	0.0375
			0.25	Project Lifecycle Understanding	0.0375
			0.5	Accounting Skills	0.075

Weighted Task consideration:

Since the hierarchy constructed already used 15 of 20 objects available to the demonstration version of Criterium Decision Plus, multiple sets of possible actions were evaluated using the S.M.A.R.T. method. This yielded a Utility value of each action, which were then recorded in an Excel spreadsheet. These results were then normed to total one, and were treated as AHP values.

The knapsack algorithm was then used to determine which tasks were most desirable given a time constraint of 10,000 hours and a budget constraint of $255,000.

Results
Objective=	.5727	Utiles
Cost=	-14	k$
Effort=	9785	hours

					Budget:	$ 255	K$
					Effort limit:	10000	hours
Option	?	Utility (SMART)	Utility (Approx. AHP)	Implementation Cost (dollars)	Implementation Effort (hours)	Integer selection variable
Prep for Degree	Y	0	0	5	200	1
Enroll in MBA	Y	0.603	0.1922	125	3600	1
Get business cert	N	0.148	0.0472	35	600	0
Get Masters OR	N	0.453	0.1444	125	3600	0
Complete UCSC PM	Y	0.148	0.0472	1	35	1
Get JD	N	0.07	0.0223	125	4500	0
Get Paralegal cert	N	0.035	0.0112	35	2000	0
Get CPA	N	0.07	0.0223	75	3000	0
Get Accounting cred	N	0.12	0.0382	10	800	0
Get job - PM oriented	N	0.445	0.1418	-150	5000	0
Bozbo.com	Y	0.223	0.0711	0	1500	1
Work with Bill	Y	0.453	0.1444	-150	3000	1
Read case studies	Y	0.148	0.0472	1	300	1
Read Nolo press	Y	0.0035	0.0011	1	450	1
Read biographies	Y	0.07	0.0223	1	300	1
Read Textbooks	Y	0.148	0.0472	2	500	1
		3.1375	1

Additionally, a number of constraints and dependencies were inputted. Degree preparation was considered as necessary for any of the degree programs, while paths that taught similar or overlapping knowledge were made exclusive.

Chapter	Risk Analysis
3

Certainty and Risk Analysis:

The prescriptions of this Analysis seem correct, and very few of the tasks which were decided upon are risk or analysis sensitive.

Before beginning this project, I had believed that a degree in Operational Research would prove more helpful than an MBA. Because an MBA is perceived as offering numerous strengths, I would have to evaluate decision making skills as being .65 Utiles in order for the Masters in OR to be the preferred course.

Most other decisions have no inflection points on comparative graphs, and are thus judged not to be risk-sensitive.

Error:

In re-examining the assumptions that were made at the outset of this project, several seem fraudulent. Were I to undertake the project again, I would emphasize the importance of various skills based on my need. Within the course of this analysis, Leadership skills were weighted as being worth as much as all other skills put together. Since I feel that I already possess a strong set of leadership skills, I would have preferred to give this category less weight in the total hierarchy.

As stated before, I am uncertain of how the SMART method is implemented. While the optimal solution would have been to purchase a full version of the software tools I was using, I could also have simply multiplied out the lower-level AHP scores by the value of the predecessors and moved the majority of the hierarchy out of the model in Criterium Decision Plus. This would have allowed me to have more lower-level factors to compare.

Plan for the future:

It is currently my intent to re-perform this analysis and post results on the silverhouse.net website. This re-analysis is intended to overcome some of the errors that were made in the pursuit of this analysis and is intended to be more complete as well. While some of the tasks that this report suggests have a high ratio of Utiles to Time, tasks such as obtaining a Paralegal certification do very poorly. Also, it may be that varying the amount of time granted and plotting these Utile results will reveal tempting local optima.