Inter-relations

[Steve Prevette]

This discussion thread is worth one extra credit point on the midterm - and may help you answer an essay question on the midterm. Please respond to this thread by October 27, but ongoing discussion is always allowed and appreciated.

In Case Problem one, we simply added up the profit contributions for the Holiday Mix, the Deluxe Mix, and the Regular Mix (of nuts). By the very nature of linear programming, we assumed no interaction between the various terms being added together in the objective function.

MBA (and project management) programs are rather typical of this - simply add together the contributions of the various components, and as long as each component does their best, all we have to do is add together these components and we will have an optimal whole!

In what cases might this be a fallacy? For example, what is the impact if the travel department attempts to optimize its costs? Might that negatively impact other departments?

Why is it that the whole rarely even equals the sum of its parts, let alone be more than the sum of the parts?

We will do an in class exercise on October 27 that will force you to examine the issue of interactions. In preparation for that, please record your thoughts here. Good luck, and have fun.

[cwoehle]

In my opinion, the main reason the whole never equals the sum of its parts is because we are rarely all working towards the same goal. Every person, every organization and every manager attempts to optimize what is in their or their organizations interest. This may at times be harmless, however, it may also result in the organization failing repeatedly to meet its goals or milestones. Every day I see a schedule that is supposed to be optimized for the company's goals, yet every day I see managers argue and scheduling change. The result of shifting priorities is always inefficiency and a less than optimal solution. Furthermore, this gives the workforce the impression that management doesnt know what it wants which results in a workforce that is working at a less than optimal pace.

[Anita Alston]

Re. the Nut Problem - so many factors were not entered into the equation:
cost of nuts on the commodities market; striking Teamsters workers and thus transportation delays from market to buyer; etc.

Re. Travel Dept. issues within corporations: I recently attended a one and a half day seminar in Boston, Ma. that took over 2 days of travel across country, with a connection in Denver, that did not factor in length of the terminal and the close connection times. Thus, the travel time was longer than the length of the seminar, and included an unexpected overnight stay.

Other, less tangible, factors that influence the "numbers": worker attitude and level of education (not everyone is created equal or produces equally), management communication skills and degree of involvement in production (which often affects workers' attitudes and productivity and can contribute to, or prevent, delays).

Personal experience: in healthcare, administrative members often want to assign a numerical value to nurses' productivity (caseload - number of patients to care for) without taking into consideration the acuity level (degree of illness and care required) of each individual patient. Thus, their productivity expectations (reflected numerically) are rarely accurate; and, often the answer is not to just hire more bodies to do the work.

Anita Alston

[Miner]

Originally Posted by Steve Prevette

In what cases might this be a fallacy? For example, what is the impact if the travel department attempts to optimize its costs? Might that negatively impact other departments?

Why is it that the whole rarely even equals the sum of its parts, let alone be more than the sum of the parts?

This is a fallacy in most cases in business. For example, Quality wants the supplier with the highest quality. Purchasing wants the supplier with the lowest cost. The best outcome for the company is the lowest cost supplier with the highest quality. Since this combination rarely exists, you make the best tradeoff between the two, preferably looking at total cost of ownership, not piece cost.

[scrowner]

linear programming can be a fallacy. The value given does not take into account of the probabilities along the way.
In the example, if one department optimizes its cost, another department may and/or will have to do without. The "probability" of keeping all departments happy is not taken into account.

Next item-for this one I would have to use the "home" for the "whole part". Hurricane Katrina-people lost their homes. All of the personal belongings are the parts to this whole home (and we treasure our personal items very much). But yet, a home in value, can be replaced monetarily, before the personal belongings (the parts). So, when I person lives in an area where hurricanes are known, they take the chance, that a hurricane "probably" won't affect them.

In other words, when we lose the parts to the "whole", replacing the "parts" can be more, than replacing the "whole". Thus, the "whole" and "parts" do not equal each other.

[terryw]

I've been thinking about the issue of inter-relations and how they can definitely affect linear programming. It's nice to have a model that tells you how things should work under perfect conditions, but competing interests and priorities can affect streamlining production and efficiencies. One example in my work is special events. We have produced performance measures to determine how many events our staff can handle and how quickly they can be done. When we developed our model, we based it on the assumption of some support items like interactive maps for organizers to use and a fully functioning software program. This hasn't happened, due to the fact that other departments, where we need the help from to produce these items have other priorities and interests. We are now forced to go down and walk each event organizer through the venue, for each event, instead of having interactive maps for them to use. This causes more staff time and throws off our production model. The same is true for our software. We have to hand carry approval papers from department to department, instead of those pieces being electronically disseminated through software. This can happen over and over again when organizational goals from department to department have competing interests.

[Jamie Morris]

Originally Posted by cwoehle

In my opinion, the main reason the whole never equals the sum of its parts is because we are rarely all working towards the same goal. Every person, every organization and every manager attempts to optimize what is in their or their organizations interest. This may at times be harmless, however, it may also result in the organization failing repeatedly to meet its goals or milestones. Every day I see a schedule that is supposed to be optimized for the company's goals, yet every day I see managers argue and scheduling change. The result of shifting priorities is always inefficiency and a less than optimal solution. Furthermore, this gives the workforce the impression that management doesnt know what it wants which results in a workforce that is working at a less than optimal pace.

I have to agree with Chris on this one. As we discussed in class, managers are usually not provided training and insight to thinking from a system perspective. Since I am a Scholtes supporter (Scholtes is also a Deming disciple), one of the new competencies that Scholtes says that managers must have is being able to view the process, operation, or organization as a system with each component or part playing a vital role in meeting the objective. To improve the process, operation, or organization, one must be able to understand the system and all its inter-related parts. In the TJ's case study that we just completed, we looked at the process from purely the linear prospective of maximizing profit. This does not take into account all the other activities (direct labor, materials, and overhead support) in the TJ's organization that is required to produce, market, sell, and distribute the three mixes to actually achieve the profit maximization. So as Chris has so aptly articulated, if managers in the other organizations in TJ's are not in tune with the profit maximization strategy, success in meeting this objective will not occur.

[jneely]

Having always worked "outside" the government contractor world, until five years ago, I feel I have a newcomers outlook. It still amazes me, even after the past five years, how the sum of the parts do not add up to the whole! In my company there are so many layers of management that the actual work and goal is often lost in the processes that are required. Procedures are often the problem. Many times these procedures have been in place for so long that they no longer fill a need or lend themselves to achieving the ultimate outcome in the most efficient manner. The many departments/levels that one must navigate often becomes the anchor that weighs us down and does not allow for success. I am in the contracts department and during our "year end" buying frenzy the department interactions swing both ways. Either we are their saviors or their demise. We either save there butts or place the noose around their necks. However, we do not have the parts to complete the "whole". Each person seems to be out for themselves and we are at their mercy. It would seem that there has to be a better way. I know that the current system is not as successful as it should be. Planning and taking the time to "touch base" with the parts will only have positive affects on the outcome.

Jo-Ann Neely MC506