Puzzles, problems, and messes in systems analysis

¶ … Puzzles - correlation to problem context

Problems contexts as social constructs

Problem analysis deems a software application to be a sort of software machine. A software development plan tries to change the problem context by producing a software machine and making it an addition to the problem context, where it will result in specific desired effects.

The specific portion of the problem context that is of attention in correlation with a particular problem -- the particular segment of the problem context that shapes the circumstance of the problem -- is named the application domain. Subsequent to the software development project has been completed, and the software machine has been placed in into the problem context, the problem context will include both the application domain and the machine (Jackson, 2005).

The problem context includes the machine and the application domain. The machine interface is where the Machine and the application domain convene and interrelate. The starting point will be to explore, in a little detail, what might be meant by the idea of a problem, especially as this is used in Systems Analysis. This discussion is based on the views of Dardenne, et al. (2004) who distinguishes clearly between problems and messes. This idea can be extended to three points on a spectrum: puzzles, problems and messes, each being intended to depict a different way in which people use the word "problem." These, however, are not the only ways - they serve as archetypes on the spectrum of possible uses.

Problems

A problem is more complicated than a puzzle, but less complicated than a mess. This complication stems from the fact that a problem has no single answer that is definitely known to be correct. At the heart of a problem is an issue that must be tackled. As they arise in real-world Management Science, an analyst is more likely to face a problem than a puzzle. An example might be, ?How many depots do we need in order to provide daily replenishment of stocks at all our supermarkets?

As a question, this seems straightforward enough and is relatively unambiguous; however, this is just the tip of the iceberg. Such a question, based on an important issue, rarely has a single correct solution, for problems, in the way that the term is used here, there may be agreement about the core issue to be tackled but there might turn out to be several, equally valid, solutions. Consider again the issue of the number of depots needed to service a set of supermarkets. One way of construing this issue might be to reformulate it as something like the following: assuming that the depots stay at their current size and the supermarkets do the same, how many depots would we need to meet current demand and to guarantee delivery to each of the stores within an agreed 30-minute time window each day?

Once formulated in this, much tighter, way there may indeed be only a single correct solution (Chan, et al. 2006).

But to achieve this, a problem, with its core issue, has been transformed (some would say reduced) to the form of a puzzle. To illustrate this point yet further, an alternative formulation of the problem might be something like: Assuming that the supermarkets and depots stay at their current sizes and are grouped into regions as now, how many depots would be needed to meet current demand in each region and to guarantee delivery to each of the stores within an agreed 30-minute time window?

Needless to say, the second formulation may produce a solution different from the first one (Dardenne. et al. 2004). There is a further reason why problems, as defined here, may lead to a range of solutions that are all acceptable, and this is to do with the approaches taken in working toward a solution. As an example, consider the type of case exercises, often known as Harvard-type cases, with which most business administration students are familiar. When used for teaching, the normal mode of use is to distribute copies of a case to the class in advance of a discussion session.

The case papers, which may be 20 or more pages long, give background on a company or market, provide extensive data and pose some issues. Later, possibly the following day, the class will assemble and, under the guidance of an instructor, students will propose approaches and solutions (Galliers, 2010). Clearly, given a range of suggested approaches, some of them will be wrong, but more than one may be plausible. There are likely to be several possible approaches, each of which may be used in sensible attempts to tackle the core issues of the case. Even when people are agreed about how to construe the basic issues, they may employ different approaches in an attempt to solve the problem. This may be due to their backgrounds, to their expertise or whatever. But there may be no real way of declaring one solution to be superior to all others, even after the removal of those that are obviously wrong (Chan, et al. 2006). This is why the case method is a good way of encouraging discussion in a class of bright students. It is also the reason why the case method of teaching business administration is preferred by some teachers who do not regard management as a generalizable science (Jackson, 2005).

Problems contexts as social constructs

What should be becoming clear from the discussion so far is that it may be best, in Systems Analysis, to regard problems, puzzles and messes as social constructs. That is, like beauty, there is a sense in which they are defined in the eye of the beholder. This does not mean that every aspect of the issues to be faced is within the mind of the analyst or the participants. That a business is heading for bankruptcy or that an ambulance service is failing to provide adequate response times may be beyond dispute (Galliers, 2010). But the interpretation of those "facts" is less certain, and different people may interpret the same "facts" in different ways. That is, there may be several valid views of what is happening and of what might be done about it. Some of these interpretations may turn out to be wrong, in the sense that they cannot be defended. However, it is very common to find more than one plausible explanation. Writing many years ago, John Dewey (quoted in Lubart, 1994) produced the maxim: a problem well put is half-solved. It seems as if he had in mind that a poorly posed problem context will be very hard, if not impossible, to solve. If this maxim is combined with the realization that problem contexts, in organizations at least, are social constructs, then the importance of problem-structuring becomes obvious.