No Title
Hammer [8], for example, argued that advances in information technology imply a wholesale restructuring of the interactions among the individual processes of the organization. In effect, though information technology has developed to solve one set of problems, the resulting technology has a much wider set of uses and the application of information technology should be directed to the most constraining business processes. The definition of business process is not always stated clearly in the literature but many contributors cite (and presumably do not disagree with) Davenport [6]. He argues that a "process approach... implies a relatively heavy emphasis on improving how work is done, in contrast to a focus on which specific products or services are delivered to customers." In other words, the opportunity costs arise from the processes undertaken within the organization and to focus on the outcomes is to focus on the wrong target.
Formal models of business process change are rejected by leading management theorists such as Weick [25] who argue that a richer language of discourse is required than formal models can provide. However, richness has costs as well as benefits. The current literature reveals many different interpretations of such widely used terms as organizational capabilities and competences. Richness is thus at the cost of imprecision in the language of discourse used to develop key concepts. A consequence of this imprecision is the difficulty of identifying independent tests of the management theories before they are put into practice with the possibility of substantial damage to the implementing organizations.
Evidently, the ability to create formal and testable models which support the same richness of discourse and applicability to (for example) business process analysis as do current approaches to business strategy would be an important advance in management science. The purpose of this paper is to define a development path which would advance the discipline significantly in that direction and to demonstrate the practicality of such a development with a detailed example on a useful scale.
The example we use highlights the difficulties organizations face in dealing with critical incidents from time to time that are not a part of their ongoing, operational routines. In some cases, these incidents can escalate into full-blown crises with catastrophic effects on the organization and its managers. A particularly compelling example of the possible effects of such incidents relates to providers of water and sewage services in the United Kingdom. Incidents involving either significant environmental damage or the supply of contaminated water can lead to gaol sentences for the company's directors in addition to the award of large damages to consumers in the courts.
One such enterprise, which has recently undergone reengineering of its business processes, is the provider of water and sewage services in the north west of England. North West Water PLC merged in 1996 with Norweb PLC, the electricity supplier for northwest England, to form United Utilities PLC. The reorganization associated with the merger changed the remit of operations managers in the regulated water and sewage activities of the company. Whereas operations managers had been expected to handle such incidents as power failures, equipment breakdowns and mains collapses independently before the reorganization, they do not now have the resources to support such independence and are expected to call on centrally provided services to contain and remedy such incidents. Such incidents occur perhaps several times a year in the operations of North West Water and, if not dealt with in a timely and effective fashion, could lead to major crises.
Because the critical incident procedures are safety critical, confidence in the implications of a computational model should not be taken lightly. Verification of the model by comparing its outputs with experience obviously cannot precede the reorganization. Nonetheless, individual components of a model can be verified against relevant experience. Moreover, validation against some well understood formal system to ensure that the model is at least logically consistent and sound would provide a basis for discussion and examination of the assumptions and relationships underlying the model.
In order to demonstrate the feasibility of this approach, a discussion of the essential elements of such validation and verification is provided in section 2 with a discussion of the implications for modelling approaches in section 3. A model of the management by North West Water of critical incident procedures is described in section 4 with the results of simulations with that model reported in section 5. The validation and verification issues associated with the model and results are explored in section 6 followed in section 7 by a discussion of the implications for future research directions.
Generated with CERN WebMaker