Articles

......................................................................................................................

Issues Involving the Creation of An

Operations Research Group

By Roberto P. Valentin

The writer is a faculty of Business Management School of Commerce, St. Scholastica's College. A prolific writer, Mr. Valentin is an active contributor to Business World publications.

As businesses strive to become more productive and more profitable or gain leadership in their field, managements resort to creating Operations Research Groups (ORG). The role of the ORG, in the words of Leonid V. Kantorovich1, Russian Nobel Prize awardee for Economics in 1975, “is to improve ‘efficiency’ in a workshop, an enterprise, or a whole branch of industry.”

This is quite a tall order considering the complex processes in large organizations. Normally, business organizations are department-based and compartmentalized; each department is concerned only with its own individual turf. Operations Research's pursuits, however, usually transcend functional boundaries. O.R. examines the entire organization and finds ways to make the various departments operate effectively and contribute to the overall efficiency of the organization.

This generally causes political conflicts, with one department outdoing others in terms of authority, implementation schemes and model ownership.

An O.R. Group is usually created when the following conditions are present:

1. Top management feels that there are too many “loose ends” that lead to inferior profits. Specific examples include numerous commitment schedules not usually met, too many (abundant) resources that remain slack, and inefficient processes or systems.

2. Complex problems are created by many factors or variables interacting in a knotty way. This is the condition that usually bogs down and discourages potential practitioners of O.R.: developing models (in terms of equations or inequalities) that capture these interactions.

3. Many solutions exist, but the one best solution cannot be readily established. The most common (mal)practice is to accept the first feasible arrangement obtained, either for budgetary reasons, ignorance or plain indolence. To the uninitiated, this first feasible arrangement satisfies all requirements anyway, and nobody is wise enough to check on him, so it is treated as the solution. But as a general rule, the first feasible solution is not necessarily optimal (or the best).

4. Supporting data are available, or at least can be derived from existing ones. The worst-case scenario is that the data needed by the ORG will not be available anywhere. Then there is the need to observe or do a survey which, an experienced practitioner knows, may be quite formidable because this takes lot of valuable time and company resources

5. An exponent of O.R. champions the introduction of O.R. methods in the organization. Usually this is a supervisor or a middle manager who has had training in Operations Research, and is very eager to apply his acquired knowledge. At other times, it may be a Board Member, who has some degree of familiarity with O.R.’s usefulness and who would want to see it implemented, much like an “experiment.”

6. Top management is supportive of the technology. This is a must for the O.R. Group to exist and, survive.

Once the go-signal is given to set up an ORG, numerous issues will consequently arise. First there will be the problem of acceptance. Some people in the organization will question the need for the ORG, and feel that its functions duplicate those of other already existing units. Others will be threatened by the scrutiny they will be subjected to. They might resist the ORG and be uncooperative when it comes to data sharing, equipment use (for validations later) and performance reviews. This is where the O.R. Group Head will have to exercise tact, dynamism and persuasive power in order to promote the group to the organization at large. It is at this time precisely where there will have to be an open, convincing support from top management.

The other issues that the ORG will face include the following:

1. Group Size. How large should the group be? In my first O.R. job, there were seven of us analysts plus one Group Head. In my subsequent O.R. job, there were five of us. The number could be as little as two (as in a local airline company) or as much as 15 or more, as in some U.S. organizations. At San Miguel Corporation, an organization where O.R. is already deeply entrenched, there are seven analysts. The group size should not be a function of the size of the organization but of the diversity of its processes.

2. Qualifications. What level of educational attainment should members have? What area of specialization should they possess? Should a master’s degree be required? In most O.R. Groups in the Philippines, the most common fields of specialization of existing members are Industrial Engineering, Mathematics, Statistics and Economics, the traditional courses. Other fields noted were: electrical engineering and the other engineering fields, computer science and agricultural economics. Some training is usually provided, but generally, a strong background in quantitative analysis, symbolic logic and computer programming should help a budding practitioner.

3. Scope of Work. Will the ORG confine its efforts purely to O.R. matters? In some instances, fulfilling a customer (internal unit) request for assistance would require only a simple computerization job. This has happened in all ORGs. As a rule, these types of requests for assistance are not turned down so as not to diminish the company’s trust in the Group. Early on, the ORG must form its own vision-mission statement, plans and programs, but these should not be held fixed. It should grow with the organization’s needs.

4. Group Name. How will the group be known? Is it as traditional as Operations Research Group/Department? Or would it have to acquire some novel or fancy tag as Decision Specialists Group, Optimal Solutions Planning Group, Portfolio Management Group, or Risk Management Group?

5. Compensation. To protect the integrity and coherence of the group, most organizations in the U.S. are giving higher than average salaries to the ORG members. In fact some are getting the highest in their position level, compared to their counterparts in the other departments within the same organization. Most of the members of ORGs are, by selection, bright, alert, have keen sense of logic and attention to detail, can equally perform analysis as synthesis, possess retentive memory and capacity to do long, arduous work. These make them very vulnerable to other departments who may lure them with higher pay. Hence the necessity for higher than average compensation. But of course, they should not be prevented from joining other departments as part of their career development in the organization. Most ORG members eventually become managers in other departments in less than three years.

6. Supervision. The decision on which officer of the company the ORG will be reporting to is vital to the success or failure of the group. In the Philippines, these are usually the vice presidents for corporate planning, finance, or manufacturing, in that order. In some organizations in the U.S., the ORG is directly attached to the office of the President. This is the most ideal scenario—the ORG will not be beholden to any major functional head. What is important is that the integrity and independence of the group should not be compromised by its affiliation with any specific division.

Some implementation problems, identified by C. P. Pappis2 that would confront the ORG include: lack of qualified personnel, insufficient data, difficulties understanding and defining the real problem, resistance by affected parties, lack of backing by top management, lack of appropriate software packages, need for quick fixes, insufficient motivation of model beneficiaries, lack of publicity of existing applications.

These implementation problems, however, could be overcome if the need for an ORG is acknowledged and the group is accepted by everyone as part of the company’s efforts to be more a successful organization.

Bibliography:

1. Leonid V. Kantorovich, Mathematical Methods in the Organization and Planning of Production, 1939

2. C. P. Pappis, G. D. Nikitas, and G. S. Patrikalakis, “O.R. in Practice: Results of a Survey in Greece”, European Journal of Operational Research, vol. 40, 1988