Portfolio > 4c > 1. Application
Search this site

4. A collaborative consultant with skills in:
(c) Problem-solving and decision-making

4c1 Application to significant technical issues

Development Plan Portfolio Documentation

Apply problem-solving and decision-making methods to significant technical issues:

  • prioritizing IT projects
  • selecting a strategically valuable groupware solution
Using decision trees, flow charts or other appropriate methods, document the process used in making significant decisions ITS.
 
  1. Decision-making Methods
    1. Flowcharting
    2. Decision Tables
    3. Kepner Tregoe
    4. de Bono's Parallel Thinking
    5. Weighted Score Rankings
    6. Analytical Hierarchical Process
  2. Prioritizing IT Projects
    1. Selecting Strategic IT Projects at Andrews University
    2. IT Project Approval Process at Sanitarium
    3. IT Portfolio Management
  3. GroupWare Selection
  4. List of References

1. Decision-making Methods

In this section, I will describe some of the decision-making methods I have used, both for making decisions in my work, and for programming decision-making capability into software I have developed.

1.1 Flowcharting

This simple example is based on one of the decision tables I used in the Student Records System I wrote in COBOL for Avondale College.  It is taken from a subroutine (named SRB$FEE_CALCULATION).  The subroutine calculates the number of credits or credit points being taken, then determines tuition, fees and discounts according to a range of criteria.

In this example, the dependents discount is determined based on the campus attended, whether the student is living on campus (indoor) or in the community (outdoor), and whether the student is a missionary dependent.

1.2 Decision Tables

Flowcharts and decision trees may be useful for representing decision flows graphically, but decision tables provide a benefit that those methods do not possess.  Decision tables provide a convenient method for ensuring complete rigor in handling every combination of conditions.

For instance, the flowchart above is clean and simple and easy to read.  However, what about the case of a missionary dependent who is staying in one of the residence halls (indoor).  According to the flowchart, they will be awarded the indoor discount.  However, this is incorrect.  The fourth column of the decision table shows that as long as a student is not studying on the Wahroonga Campus, if they are the dependent of a missionary, they will get the missionary discount regardless of whether they live in the residence hall (indoor) or in the community (outdoor).

1164  /
1165  D2810_DEP_DISCOUNT SECTION.
1166  
1167  ********************************************************************************
1168  * FUNCTION: Compute Dependents Discount, rounded to nearest 10 cents
1169  ********************************************************************************
1170  
1171  D2811_DEP_DISCOUNT.
1172  
1173  * +-----------------------------------------------+
1174  * |Dependents Discount for requested semester     |
1175  * |DEPNT_DISC_PCENT                               |
1176  * |-----------------------------------------------|
1177  * |Wahroonga Campus               | Y | N | N | N |
1178  * |Indoor Dependent               | - | Y | N | - |
1179  * |Outdoor Dependent              | - | N | Y | - |
1180  * |Missionary Dependent           | - | N | N | Y |
1181  * |===============================================+
1182  * |P_FE_DEPNT_DISC_I_PCENT        | X | X | - | - |
1183  * |P_FE_DEPNT_DISC_O_PCENT        | - | - | X | - |
1184  * |P_FE_DEPNT_DISC_MISNRY_PCENT   | - | - | - | X |
1185  * +-----------------------------------------------+

The error made in the simple but incorrect flowchart implementation referred to above could have been avoided if the decision table had followed one of the principles of decision table design.  Columns with more "don't care" conditions should appear in the table ahead of columns with fewer such conditions.  Hence, column 4 should be moved to column 2.  At the same time, we should probably move condition row 4 to row 2.

1173  * +-----------------------------------------------+
1174  * |Dependents Discount for requested semester     |
1175  * |DEPNT_DISC_PCENT                               |
1176  * |-----------------------------------------------|
1177  * |Wahroonga Campus               | Y | N | N | N |
1180  * |Missionary Dependent           | - | Y | N | N |
1178  * |Indoor Dependent               | - | - | Y | N |
1179  * |Outdoor Dependent              | - | - | N | Y |
1181  * |===============================================+
1182  * |P_FE_DEPNT_DISC_I_PCENT        | X | - | - | - |
1183  * |P_FE_DEPNT_DISC_O_PCENT        | - | - | X | - |
1184  * |P_FE_DEPNT_DISC_MISNRY_PCENT   | - | X | - | X |
1185  * +-----------------------------------------------+

In this rather simple example, there are four independent conditions and three possible outcomes.  The four conditions could lead to a maximum of 16 different outcomes, but as the decision table shows, if a student is on the Wahroonga Campus, the other three conditions do not matter.

This decision table occurs at line 1173 in the subroutine.  It is implemented using nested IF statements, which can be quite confusing to read.  There is another one (very simple) at line 904 and a more complex one at line 793.  This one is implemented using COBOL's EVALUATE statement (commonly referred to as a "CASE construct"), which is a much more direct implementation of the decision table, and is easier for verifying correctness.  In fact, this close correspondence between the representation of the decision logic and a computer implementation has led to the creation of software for automating the process.  One example of such software is Cope.  Barry Dwyer, a senior lecture in Computer Science at the University of Adelaide in South Australia, wrote Cope and has an online manual for using the software.  He writes in his introduction to the manual,

"Cope is a program that converts decision tables to Cobol source statements. This text assumes that you have some knowledge of Cobol, but it does not assume any knowledge of decision tables.

"A decision table is a way of expressing logic in tabular form. It comprises three elements: a set of conditions that control the logic, a set of executable actions, which depend on the conditions, and a set of rules, which interrelate the conditions with the actions. The rules specify the paths of logical flow through the program.

"A decision table performs a similar function to a flowchart. The two are alternatives, but a table has the advantage of being better suited for computer processing. The relationship between flowcharts and decision tables is explained in more detail in the section Converting Flowcharts to Decision Tables." Barry Dwyer

Marien de Wilde (2002) has written an excellent PowerPoint presentation describing decision tables and how to write them.

1.3 Kepner Tregoe

While working at Hewlett-Packard, I attended many workshops and seminars in personal and professional development, such as "Working at HP", giving presentations, time management, and problem solving.  In one of these seminars, I was trained in the Kepner Tregoe method of problem solving.  Charles Kepner and Benjamin Tregoe (1981) developed their rational approach to problem solving and decision making in The New Rational Manager.  I learned a very simple but effective first approach for solving the kinds of problems I would be most likely to meet in my position in IT support.  These two questions often reveal causes that are easily overlooked.
  1. Has this feature or process worked correctly in the past?
  2. What has changed between then and when it stopped working?

Attendees at the Kepner Tregoe seminar were given a little two-sided plaque in a stand to place on their desks, summarizing the essential elements of the "rational process" for resolving problems.  I have scanned both sides and included them in this document below.

Note: Another technique that is often used for discovering root causes is the Cause and Effect method.

1.4 de Bono's Parallel Thinking

In Six Thinking Hats, Edward de Bono (1985) describes parallel thinking as opposed to argumentative, confrontational, and adversarial thinking. Too often, we make decisions in committees based on the best argument, where each party speaks only in favor of its point of view. "If one party thinks of a point that might benefit the other party, then that point is never raised. The purpose is to win, not to explore the subject honestly." p9

De Bono uses the mechanism of six colored hats, with each hat representing a different style of thinking. The aim of this method of parallel thinking is to have everyone thinking from the same perspective at a time, say wearing the White Hat. Then after a period of time, everyone in the committee 'puts on' the next color hat. De Bono maintains that there is an "absolute physiological need to separate out the types of thinking ... You cannot be sensitized in different directions at the same time, so when we set out to do all aspects of thinking at the same moment, we are going to be suboptimal on all of them." p12

Parallel Thinking using de Bono's Six Thinking Hats
White Hat Neutral, objective.  The white hat is concerned with objective facts and figures.
Red Hat Suggests anger (seeing red), rage, and emotions.  The red hat gives the emotional view.
Black Hat Somber and serious.  The black hat is cautious and careful.  It points out the weaknesses in an idea.
Yellow Hat Sunny and positive.  The yellow hat is optimistic and covers hope and positive thinking.
Green Hat Grass, vegetation, and abundant, fertile growth.  The green hat indicates creativity and new ideas.
Blue Hat Cool, the color of the sky, which is above everything else.  The blue hat is concerned with control, the organization of the thinking process, and the use of other hats.

1.5 Weighted Score Rankings

I have used this method to rate different vendors when purchasing a large server, and on occasion I have used it to rank candidates for a vacant position, but only as one part of the decision-making process.  In an MIS (Management Information Systems) course I have taught a number of times, I teach this method from Laudon and Laudon's (2002) Management Information Systems textbook, chapter 11.

In the weighted scoring model, there are a number of alternatives from which to choose one or more that best meet certain criteria.  I will summarize briefly the steps that are followed.

  1. Identify desirable features or criteria, as well as risk factors.
    Each alternative will be rated against these desirable features or criteria.
  2. Determine relative weights for each of the criteria.
    The criteria do not all have equal importance.
    Typically, the weights will add up to 1.00.
  3. Score each alternative for each of the criteria.
    Some of the criteria will be expressed in numbers (monetary, time, etc) and can be scored somewhat objectively.
    For others, assigning a numerical score will be much more subjective.
  4. Multiply the scores by the weights, sum, and rank-order.
  5. Select the highest ranked option(s).

I set an assignment using weighted score rankings which serves as a good illustration of the method.

1.6 Analytical Hierarchical Process

One of the sessions at the 2002 Leadership Roundtable Conference was called "Decision Making for Leaders" and was facilitated by Janusz Kobielski.  It was an excellent session, and Janusz brought as his special guest an acknowledged leader in decision-making process and technology, Dr. Ernest H. Forman.  Dr. Forman is a Professor of Management Science at George Washington University, and is co-founder of Expert Choice Inc.  ExpertChoice is a computerized implementation of the Analytic Hierarchy Process (AHP), a methodology for group decision making and prioritizing alternatives.

Frank Moisiadis (2002) presented a research paper at the October 2002 Test & Evaluation Conference in Sydney Australia entitled "Fundamentals of Prioritizing Requirements".  In his study of the prioritization process, he does a critical analysis of current techniques such as AHP (Analytic Hierarchy Process) and QFD (Quality Function Deployment).  See pages two and three of his paper.

The AHP methodology starts out in a way that is similar to the weighted score rankings method described above--identify the alternatives that are being prioritized, then decide on the criteria to measure them all against.  In the AHP method, you have clusters of nested criteria, and the mathematics handles the weighting as the scores are "rolled up".  But it is the scoring process itself that is very different in AHP.  Instead of having the participants rate each alternative on a scale of one to 10, the scoring is computed from pair-wise comparisons.  For each criterion, the alternatives are compared with each other in all the possible combinations of two at a time.  Mike Raisinghani and Lawrence Schkade have a good summary of the method in the proposal page of their study into Strategic Decision Making, under the heading "Strategic Justification Methodology".

The person responsible for devising AHP is Thomas Saaty.  This was in the 1970s.  In 1983, Saaty's consultancy incorporated as Expert Choice when  Ernest Forman patented the AHP decision support software for PCs.  Forman is now president of Expert Choice.

The ExpertChoice.com web site claims the following benefits (among others), for the Expert Choice method:

I downloaded a trial version of the software, and was really impressed with the ease of use of the user interface, and how it leads you through the evaluation process.  I was also impressed with the high power mathematics that handles inconsistencies (e.g., if I score A more than B and B more than C, if I score C more than A, then that is inconsistent).  Using linear programming techniques from Operational Research, the software is able to arrive at the set of scores that is the best fit given the inconsistencies.

I tried a very simplified version of the method with 12 IT projects/technologies at Sanitarium in a spreadsheet.  I do not have multiple criteria for scoring each of the 12 alternatives.  In a pair wise fashion, I merely rated each alternatives importance, in my mind, relative to the importance of each of the other alternatives.  The end result in this simple test is approximately what my "gut feel" tells me.

2. Prioritizing IT Projects

2.1 Selecting Strategic IT Projects at Andrews University

In 1997, I organized and hosted the MIS Directors Conference.  In preparation for the conference, I conducted survey asking potential attendees to vote on a list of IT issues to select those they would most like to see on the agenda.  I highlighted the top seven from the list of 25 topics.

During 1999 at Andrews University, ITS compiled a list of information technology projects that were being requested by various groups around campus.  Personnel within ITS wrote one-page summaries for each of the 15 strategic information technologies that were identified.  We took votes from the three University computing committees, the IT management team, the Strategic Planning Committee, and the President's Cabinet.  The voting process is described and the results are summarized in it_project_voting.htm

In June 2001, John Jasinski led a group of Andrews University administrators in a workshop entitled "Performance Excellence at Andrews University".  A number of recommendations were made, based on Baldridge Criteria.  One thing that impressed me was the number of references to fact-based decision making.  This was taken as strong support to press on with plans to implement data warehousing at Andrews.

2.2 IT Project Approval Process at Sanitarium

On joining Sanitarium Health Food Company, I soon discovered that the pressures on IT were much the same in the corporate world as I had them to be in my 15 years in IT in higher education.  Requests come to IT from all over the company, and there is quite a backlog of project requests.  This leads to departments looking for their own IT solutions, without the benefit of strategic appraisal at the executive level.  To address this potential for poor IT investment, we have been working with key players to develop an IT Project Approval Process that will be in everyone's interest to follow.  We have depicted the process as a flowchart with two bands - one for the Business Area, and one for the IT Steering Committee.  The flowchart illustrates several key points:

After presenting a draft of our proposed approval flowchart to senior administrators at Sanitarium, I read an article by Paul Smith (2003) in the April 2003 issue of MIS - Managing Information Strategies which addressed the very issue of the approval process for IT projects.  Colonial First State have a two-phase approval process just as we were proposing.


Colonial First State's 2-phase decision-making structure

ClickStream's decision flow was represented in another diagram as a cross-functional flowchart just as we did, but they broke it down to lower levels of detail, and had five bands rather than our two.  This diagram also made it clear that the identification of need and the creation of a business justification are the responsibility of the business area, and that this must be supported by a powerful sponsor.  The diagram also relates all requests back to business strategy as implemented through IT strategy.

CLICKSTREAM'S BEST PRACTICE I.T. DECISION FLOW
Source: ClickStream
ClickStream 's multi-band approach to decision making

2.3 IT Portfolio Management

In the 2003-2004 budget process at Sanitarium (my first in the new job), a variety of pressures were coming to bear on IT resources.  The network infrastructure for the enterprise was aging, and plans needed to be put in place for its replacement.  And business units were wanting to make more sophisticated use of the Enterprise Resource Planning (ERP) system the company had invested in by implementing more of its unused potential.  Altogether, these pressures would require a significant multi-year investment.  So it was clear that project approvals and prioritization needed to happen at the executive level within the company, and I knew it was my responsibility to prepare an overview of what needed to be done.

As I looked for ways to present the needs of the company to senior management, I came across references to an approach called "IT Portfolio Management".  This approach has its roots in Modern Portfolio Theory (MPT), which "describes how, for a given risk level, there is a specific mix of investments that will achieve an optimal return." (A Summary of First Practices and Lessons Learned in Information Technology Portfolio Management, page 2).  IT portfolio management is now regarded as as one of the "central elements of good IT investment management." (page 3).

In this 39-page report prepared by the US Federal CIO Council Best Practices Committee, IT portfolio management is defined as follows:

"In the simplest and most practical terms, portfolio management is about the five following items:

  1. Defining goals and objectives – clearly articulate what the portfolio is expected to achieve.
  2. Understanding, accepting, and making tradeoffs – determine how much to invest in one thing as opposed to something else.
  3. Identifying, eliminating, minimizing, and diversifying risk – select a mix of investments that will avoid undue risk, will not exceed acceptable risk tolerance levels, and will spread risks across projects and initiatives to minimize adverse impacts.
  4. Monitoring portfolio performance – understand the progress that the portfolio is making toward the achievement of the goals and objectives.
  5. Achieving a desired objective – have the confidence that the desired outcome will likely be achieved given the aggregate of investments that are made." (page 4).

This document makes many helpful points, and lists nine lessons learned from experiences in the US federal government.

Lesson 1. Understand the differences and the relationship between portfolio management and project management and manage each one accordingly

Lesson 2. Gain and sustain the commitment of Agency officials and senior managers to make informed IT investment decisions at an enterprise level and to uphold them

Lesson 3. Establish and maintain an enterprise architecture to support and substantiate IT investment decisions

Lesson 4. Integrate IT portfolio management with the organization’s planning and budgeting policies, processes, and practices.

Lesson 5. Clearly define and communicate the goals and objectives to be served by the IT portfolio and the criteria and conditions for portfolio selection.

Lesson 6. Acquire and utilize portfolio, project management, decision support, and collaborative methodologies and tools

Lesson 7. Routinely collect and analyze data and information to assess portfolio performance and make adjustments, as necessary

Lesson 8. Carefully consider the internal and external customers and stakeholders of the organization’s IT portfolio

Lesson 9. Pay very close attention to the inter-organizational aspects of the organization’s IT portfolio

I located another useful web site maintained by the Washington State Department of Information Services.  This site was a big help as I began drafting my first Sanitarium IT Portfolio.

3. Groupware Selection

Andrews University gave a lot of study to the question of what groupware software to use to "white collar productivity" and collaboration.  When I joined the University in 1996, Novell was the network operating system that was used for file and print, WordPerfect was well established for word processing, and most people used a character-based Unix program for email.  The options we considered for a groupware solution were:

Initially, it made a lot of sense to go with Novell's GroupWise, since we were already using their networking software.  But even the strong Novell advocates in ITS agreed that would not be a good choice for Andrews.  So that left three options.  The functionality we were looking was not yet available as an Open Source solution, so that left Microsoft and Lotus.  We actually had a consultant come and spend a day with us to help us decide between the two.  In the end, the decision was made to go with Microsoft, largely on the basis that it had more functionality straight out of the box than Lotus, and therefore could be deployed with lower support costs.

Over the next six years, Microsoft Excel came to be the dominant spreadsheet software that was used, and most people used Microsoft PowerPoint for presentations, and Microsoft Access for personal database applications.  Outlook was beginning to be the preferred email and personal information manager by users.  ITS had made the decision to centralize logon authentication and access rights into Microsoft's Active Directory, and to switch from Novell networking to Microsoft Windows networking.  In almost every way, Andrews had become a Microsoft campus.  The one exception was in word processing.  The official standard continued to be WordPerfect, although, through sales and support data in ITS, it was apparent to us that that change was taking place unofficially as students and new faculty and staff tended to be Microsoft Word users.

When I raised the issue of a standard for computer office suite software in 2000, it was suggested that while the corporate world might be moving to Microsoft, this did not seem to be happening in academia.  So I sent a very brief email questionnaire to the EDUCAUSE CIO listserv.  Altogether, I received and tabulated 29 responses.  Three of the 29 supported WordPerfect (two of them jointly with Microsoft Word).  Only one of the 29 did not support Microsoft Word, but were thinking of making the move (see summary).  Colleen Keller, Information Specialist for EDUCAUSE, requested permission to post a link to the survey on the EDUCAUSE Information Resources page (http://www.educause.edu/645?PARENT_ID=375), which I was glad to give.

It seemed to me that if I were to force a vote on the issue of WordPerfect or Microsoft Word, that personal preference, experience, and familiarity would favor stating with WordPerfect regardless of the trends in business, the rest of academia, and even on our own campus, so I chose to let the matter take its own natural course for the time being.

In mid-2002, a statement made by a student PC support technician was interpreted to mean that ITS had unilaterally decided to cease supporting WordPerfect.  So I knew it was time to approach the whole question of which office suite and which word processor should Andrews use.  I decided to conduct a survey of current usage to determine what training seminars and workshops would be the most appreciated.  It was interesting that during preliminary discussions and meetings with faculty groups and Deans' Council, I was given two seemingly impregnable arguments against leaving WordPerfect.  Since Andrews was a General Conference (GC) institution, and the GC standard was WordPerfect, Andrews should use WordPerfect.  The truth was that the GC had made Microsoft Word their standard 18 months earlier (plus, it Andrews is not bound to follow the GC anyway).  The other argument was that Andrews' affiliate institutions use WordPerfect and can't afford to switch to Word.  The truth in this case turned out to be the complete reverse - they are already using Word.

Anyhow, I conducted the survey, and received 205 responses.  There were slightly more installations of Word than of WordPerfect.  I wrote up the whole experience for competency 6(b) Theories of learning and human development as an "Application to a real life case study".

Most of the data was collected and analyzed after I had accepted a call back to Australia to a position at Sanitarium Health Food Company, and there simply was not sufficient time to bring the matter to a successful conclusion.  But subsequent to my leaving Andrews, some financial considerations have made the decision to move to Microsoft more compelling, and I believe that new licensing arrangements are being worked out for the Microsoft Office Suite for all students, faculty and staff.

4. List of References

de Bono, Edward. (1985). Six Thinking Hats. (EN-0141)
Ringwood, Australia: Penguin Books. ISBN: 014013784X

de Wilde, Marien. (2002). Decision Tables [PowerPoint]. (EN-0751)
Retrieved 18-Jun- 2004
URL: http://www.sqnz.org.nz/documents/Decision%20Table%20training%20session.ppt (on local server)

Dwyer, Barry. How to Write Cope Decision Tables [Web]. (EN-0750)
Retrieved 18-Jun- 2004
URL: http://www.cs.adelaide.edu.au/users/dwyer/COPE-MAN.html

Federal CIO Council Best Practices Committee. (2002). A Summary of First Practices and Lessons Learned in Information Technology Portfolio Management [pdf]. (EN-0752)
Retrieved April 12, 2003 (not found on 18-Jun-2004)
URL: http://www.cio.gov/archive/BPC_portfolio_final.pdf (on local server)

Forman, Ernest H. (2003). Expert Choice [Web]. (EN-0744)
Retrieved 18-Jun- 2004
URL: http://www.expertchoice.com/

Kepner, Charles H. and Tregoe, Benjamin B. (1981). The New Rational Manager. (EN-0041)
Princeton: Princeton Research Press. ISBN: 8084367
Retrieved 18-Jun- 2004
URL: http://www.kepner-tregoe.com/

Laudon, Kenneth C. and Laudon, Jane Price. (2002). Management Information Systems: Managing the Digital Firm (7th ed.). (EN-0745)
Upper Saddle River, NJ: Prentice-Hall Inc. ISBN: 0130330663 (editorial reviews)

Moisiadis, Frank. (2002, October). The Fundamentals of Prioritizing Requirements. (EN-0762)
Paper presented at the Test & Evaluation Conference,, Sydney, Australia.
Retrieved 18-Jun- 2004
URL: http://www.seecforum.unisa.edu.au/SETE2002/ProceedingsDocs/05P-Moisiadis.pdf
(local server)

Raisinghani, Mahesh (Mike) S. and Schkade, Lawrence L. Strategic Decision Making: A Framework For Multicriteria Decision Analysis Of Technology Investments And A Field Survey [Web]. (EN-0763)
Retrieved May 18, 2003 (not found on 18-Jun-2004)
URL: http://hsb.baylor.edu/ramsower/ais.ac.97/papers/raising2.htm

Smith, Paul. (2003, April). Cooking With Gas: How To Make Your I.T. Projects Sizzle. Managing Information Strategies, p42. (EN-0754)
Retrieved 18-Jun- 2004
URL: http://www.misweb.com/magarticle.asp?doc_id=21605&rgid=2&listed_months=0
(on local server)

Washington State Department of Information Services. (2002, April 2002). Information Technology Portfolio Management Standards [Web]. (EN-0753)
Retrieved 18-Jun- 2004
URL: http://www.isb.wa.gov/policies/portfolio/101S.doc


Return to 4(c) Problem-solving and decision-making

Created: Sunday, February 20, 2000 05:43 PM
Last Modified: Thursday, October 6, 2005 12:45 PM