This paper presents an account of a study undertaken jointly by the University of Canterbury and Tait Electronics Limited, a leading New Zealand manufacturer of high-technology radio communications equipment. The project was part of a wider research programme exploring design management and the uptake of computer aided engineering tools in New Zealand industry, funded by the New Zealand Foundation for Research Science and Technology. The study reported in this paper investigated techniques by which the product development cycle may be reduced, and in particular, the practical implementation of concurrent engineering principles in the company's product development environment. The research methodology involved the application of design phase diagrams and the recording of instances of communication between intracompany groups and departments, and organizations external to the company. The study highlighted strengths and weaknesses in the company's design management. Particular weaknesses were the level of communications between engineering and marketing functions, and recurrences of task clarification and conceptual design activity late in the overall product development process because of incomplete definition of customer design requirements at the design proposal stage. It was concluded that, with automated data gathering, design phase diagrams could provide a valuable design management tool to evaluate ongoing project progress, and to identify weaknesses in communication or design team effectiveness.

A graph theoretical approach is presented that permits the calculation of optimal process plans from datum hierarchy trees, which are elegant representations of tolerance charts. The theory is developed for a cost function that minimizes machine and datum changes. Restrictions on the sequence of machining operations arising from technical constraints or the layout of a group cell are taken into account. The process plans are optimal with respect to a specified datum hierarchy tree, manufacturing constraints and cost function. It is shown that the number of possible plans for real industrial trees is very large, and therefore this kind of optimization is beyond the ability of a human process planner. Trial runs with industrial parts indicate that optimization can be performed by a computer within a time period that is acceptable for industrial use.

Commonly used models of the design process do not reflect concurrent engineering, the influence of computer-assisted design on the design process, or the recursive divergent-convergent thinking processes at different levels of detail from whole system to subsystem to component design. Also, these models do not illustrate the process of establishing functional integrity in the design of multicomponent systems where multiple interdependencies are present. This paper outlines design model developments by the authors and introduces a methodology for designing for system integrity and minimizing risk from early in the conceptual phase, where uncertainty is high. The process involves probabilistic reasoning in propagating both qualitative and quantitative effects of changes or uncertainty in component or subsystem specifications through system models and determining the integrity of the design from multiple viewpoints.

This paper describes the development of Computer Aided Process Planning methods at the University of Canterbury and Nanyang Technological University. Three methods are described: a generative system for milling operations, an interactive system for cylindrical parts, and a retrieval system using neural networks. They are based on tolerance analysis and, in particular our technique for tolerance charting using rooted tree digraphs.

This paper outlines an applied research programme to automate tolerance charting using graph theory. The programme consists of two main phases: 1) developing a rooted tree technique for tolerance charting; and 2) developing parametrized tolerance charts. The first phase, which has been completed, involves the development of a rooted tree technique for producing tolerance charts and its implementation in an industrial software package called CATCH. The rooted tree technique is explained and illustrated with a simple example. The second phase, which has recently commenced, aims at trying to parametrize tolerance charts. The output of the second phase is intended to be a neural network that can accept the part geometry as input and output a "skeleton rooted tree". A preliminary project has been carried out to determine the feasibility of classifying trees with a neural network according to industrial practice in one manufacturing company. The results of the feasibility study are presented.

SUMMARY This paper examines the management challenges encountered in running undergraduate projects in collaboration with industrial partners. A cooperative project between the University of Canterbury, Christchurch, New Zealand and Feltex Woven Carpets Ltd to design an automated machine to punch coding cards for Jacquard looms is used as an example. The design work described includes the mechanical system, control electronics, and communications hardware and software. The new machine is now in full-time service and performing very reliably.

This paper argues that tolerance charting is not relevant for design tolerancing, but that it is an effective tool for process planning. It also argues that the tolerance charting algorithm based on rooted tree graphs, developed by the authors, provides a very general and powerful representation technique for tolerance charting. The technique is explained in detail to show how it can encompass all practical manufacturing processes that affect dimensions. A computer-based implementation of this technique, called CATCH, has been developed and is currently being used by a company in Singapore for process planning.

This paper describes a cooperative project between the University of Canterbury and Feltex Woven Carpets Ltd, Christchurch, to redesign an existing unreliable automated machine for punching coding cards for Jacquard Gripper weaving looms. Design and development work reported in the paper was carried out over three years and included the mechanical system, control electronics, communications hardware, and software. Aspects of the management of industry-based design in undergraduate student projects are also discussed,

The paper explains the implementation of the concept of workpiece control, in particular dimensional and geometric control, in planning the machining sequence and in designing location systems for prismatic parts with three types of machined features: plane surfaces, steps and slots. The system developed is capable of generating the required process properties for all machining operations and incorporates a tolerance charting facility for analysing all dimensions produced by the process. Though at present the system is limited in capability, it is capable of development into a practical CAPP system.