4.10 Concluding Remarks

The assessments made in chapter are summarized in figure 4.2. There are several conclusions to be drawn.

First of all, there are many areas of research and practice that use enterprise modeling actively as a technique and thereby represent a potential source to draw upon in further research. Still, the areas expose differences in both purpose of modeling and type of work that is modeled, making them not necessarily directly comparable.

Secondly, there are more areas that focus on modeling of replication risk than on modeling of work associated with design risk. This should not be surprising, as representation of work that are not well known necessarily have to be more difficult than routine work. Computer support is first sought in areas that represent the least challenge.

A third remark is that there is a dependency between type of work and purpose of modeling that do not show up in this superficial assessment. Not all types of work make themselves readily available for automated support, e.g., creative work is not modeled for automation of the process.

Figure 4.2: Categorization of some areas relevant to enterprise modeling

A fourth remark, although of less importance in this thesis, is that several of the areas overlap considerably in research issues and approaches. Hence, cross-disciplinary research might be beneficial. Certainly, there are several claims of the same, e.g., on the relationships between SE and ISE (Conradi, Høydalsvik and Sindre, 1994), CE and CIM (Kovács, Mezgár and Nacsa, 1994), CE and SE (Aoyama, 1993), BPR and WM (Ramage, 1994), and BPR and SE (Warboys, 1995).

Compared with the three initial empirical studies, the most promising areas that fit the perspective on modeling taken in this thesis are ISE, BPR and to some degree CE. The main reasons are their focus on modeling primarily for humans sense-making and communication, and allowing models to be informal in the sense that they are not required to be computational. The qualitative part of SD is also interesting, although the quantitative aspect caused a lot of problems in the Gazz project.

A final remark can be made if revisiting section 1.2.2: The essence of chapter 4 is that the answer to the main research question will vary depending on the area in which the modeling effort is intended to take place. The requirements to an approach to enterprise modeling within CIM might be very different from that within ISE, as the models may be used differently and the modeling processes may pose different challenges.