----- Begin Included Message -----
Date: Tue, 18 May 93 16:58:13 CDT From: jl@uhics.ics.hawaii.edu
-------------------------------------------------------------------------------
The following description of the Process Handbook project should be relevant to the recent discussion on the connection between business process reengineering and design rationales. This project includes participants from MIT, Univ. of Michigan, Univ. of Hawaii, and UCLA. (Malone, Crowston, Lee, and Pentland 1993).
The goal of the process handbook project is to support process analysis and redesign by creating an electronic handbook organizing numerous example processes and an environment for systematically exploring various combination of these processes. The processes are represented and organized along two major relations -- decomposition, specialization -- as well as many other dependency relations such as prerequisite, producer-consumer dependencies.
The process handbook uses these relations to help the user better understand existing processes and explore alternatives systematically. Because a process is represented as a specialization of more general processes, the user can better understand what it shares with other similar processes and how it is different. Because the decomposition of a process tells the user about its subprocesses, each of which is itself a specialization of other processes, the user can systematically generate and explore alternative process designs by trying out different specializations for each of the steps.
Process descriptions are being collected from field site companies, from case studies, and from other relevant literature . In each of these cases, we will attempt to extract the rationales underlying the processes and systematically associate them with each handbook entry. For example, if Boeing used a blackboard-based communication process to coordinate its assembly operations, we will try to find out why it was chosen and what alternatiives were explored. We will also study, when possible, what the consequences were. These rationales will then be used to help explore new alternatives as well as better understand exsiting processes. The representation of these rationales will be based on DRL (Lee'92), semi-formally represented initially but incrementally formalizable as the needs arise.
A major challenge is to separate out the credit assignment problem, that is how can we tell that which choice of processes, if at all, resulted in the successful or disastrous outcome, as the case may be. I have been looking at planning literature for insights, and the recent message by A.Tate on plan rationales seems very interesting. I would be very interested in forming or joining a subgroup discussing these topics.
Lee, J. & K.-Y. Lai (1991). What's in Design Rationale? invited submission to Human-Computer Interaction special issue on design rationale 6(3-4) pp. 251-280
Malone, T. W., Crowston, K., Lee, J. and Pentland, B. Tools for inventing organizations: Toward a handbook of organizational processes. Proceedings of the 2nd IEEE Workshop on Enabling Technologies Infrastructure for Collaborative Enterprises, Morgantown, WV, April 20-22, 1993.
----- End Included Message -----