An Active Information Systems Approach ToIndustrial Planning
Graham Winstanley · WIT transactions on information and communication technologies · 1970
Industrial planning involves an in-depth understanding and continual consideration of all phases in the product lifecycle, and the interdependences of sub-tasks, from conceptual design to implementation and beyond. In some engineering disciplines; for example construction, the problem is compounded by the number and types of disparate disciplines involved in the lifecycle, and in the case of major projects, the final results are all too often characterised by budget deficits and project timescale overruns. This paper describes an approach to industrial planning which relies on a centralised symbolic model of the proposed project in terms of a fusion of three distinct relational models; component, action and resource. The paper describes the philosophy of modelbased planning and its application in the industrial context. 1.0 The problems of industrial planning Planning is not a trivial process, and conventional computer-based tools have satisfied only a small fraction of the needs of industrial users. The process involves an in-depth consideration of the entire lifecycle, and this can only be achieved through a thorough and up-to-date understanding, not only of the various sub-tasks in the process, but also an appreciation of their sometimescomplex interdependences. 'Conventional' tools are very efficient when used to present such information, and the best of them are designed to facilitate a good deal of project manipulation and experimentation through powerful userinteraction tools. However, the user of such systems has the responsibility of initially specifying the task-level project, based on personal understanding and expertise. It would assist greatly if the computer-based planning tool itself were able to automatically generate initial plans, based on specifications or product designs, and in addition were able to give advice on implications of design, production or specification/constraint changes. This paper describes two pieces of highly-related research and development in the area of model-based planning. The first deals with the problems of industrial design as a configuration problem, and the second is an example of planning subsequent to detailed