Michel T. Martinez

Product Family Manufacturing Plan Generation and Classification

In the concurrent engineering approach, the problems related to product fabrication are taken into account as early as possi ble in design process. Manufacturing and assembly process design are decisive and expensive processes conditioning all the future steps of industrialisation and fabrication. In this paper a new method for assembly plan generation, based on the concept of meta-prod uct, is presented. This method is designed for modern production environments that are organised in product families with multiple op tions and variants. Facing the combinatory explosion problem, the method proposes an original representation of assembly sequences allowing the generation of a minimal list of non-redundant assembly plans called meta-sequences. In this method, an optimal assembly sequence class is selected. Then, for each product variant an optimal specific plan class is deduced from a simple analysis of product variants components. An assembly reactive control system, adapting this optimal solution to the real assembly workshop conditions is also proposed. Another competitive problem is the case of production of old or short-series products. To rationalise their manufacturing, a solution consists in classifying products into classes, on the basis of similarity of required manufacturing processes. This technique called group technology (GT) helps to minimize unnecessary variety of products in a manufacturing plant by making designers aware of existing simi lar components. In this paper a new GT method is presented. It allows the classification of sequential manufacturing sequences as well as tree-like manufacturing plans. The method is based on an adapted representation of tree-like manufacturing plans, the distance adapted to this special case, efficient heuristics, and hierarchical bottom-up classification. Then, each product class can be managed in accordance to our family product generation method, or can be served as a base for reengineering process of firms range of products.

Dynamic generation of disassembly sequences

Disassembly for material recovery and environment protection is an increasingly important aspect in the manufacturing industry. Most of the present research works for disassembly focus on the generation of predictive optimal sequences. In fact, the disassembly process is particularly uncertain. Frequent incidents make a predetermined plan unrealistic. In this paper we present a new disassembly method in which the actual sequence of disassembly operations is progressively elaborated during the execution, in function of the real circumstances of the process. New concepts of disassembly operation and sequence equivalence are proposed together with a new representation of disassembly sequence. A decentralised and distributed control system based on a multi-agent technology is presented. The results of experimentation verify system flexibility and reactivity to unpredictable events, particularly disassembly operation failures.

Optimal assembly plan generation: a simplifying approach

The main difficulty in the overall process of optimal assembly plan generation is the great number of different ways to assemble a product (typically thousands of solutions). This problem confines the application of most existing automated planning methods to products composed of only a limited number of components. The presented method of assembly plan generation belongs to the approach called “disassembly” and is founded on a new representation of the assembly process, with introduction of a new concept, the equivalence of binary trees. This representation allows to generate the minimal list of all non-redundant (really different) assembly plans. Plan generation is directed by assembly operation constraints and plan-level performance criteria. The method was tested for various assembly applications and compared to other generation approaches. Results show a great reduction in the combinatorial explosion of the number of plans. Therefore, this simplifying approach of assembly sequence modeling allows to handle more complex products with a large number of parts.

Evaluation and exploitation of knowledge robustness in knowledge-based systems

Abstract Industrial knowledge is complex, difficult to formalize and very dynamic by reason of the continuous development of techniques and technologies. Verification and validation of knowledge bases at the time of theirs elaborations is not sufficient. To be exploitable, knowledge must then be able to be used under conditions (slightly) different from conditions in which it was formalized. So, it becomes vital for the company to permanently evaluate the quality of the industrial knowledge implemented in the system. This evaluation is founded on the concept of robustness of knowledge formalized by conceptual graphs. The evaluation method is supported by a computerized tool.

Product and process modelling in a cooperative environment: a case study for thermal power plant design

Today, the product development approaches are evolving rapidly. Collaboration among multiple organisations and enterprises becomes an important key factor for design and construction of large and complex systems. To facilitate this collaboration, it is necessary to create an environment that enables integrated product and process development. The objective of this paper is to explain a distributed information infrastructure, which shares product data and process models in a collaborative environment. For this purpose, we firstly describe the problems related to product development and present a summary of actual research works in this domain. We represent the concepts related to product development using STEP standard. Then the proposed cooperative architecture is presented. Finally, we explain how we can design collaboratively a thermal power plant using these models.

Data architecture for collaborative conceptual design

Currently, computer aided systems have concentrated on the capture and representation of geometric shape and technical information, as opposed to providing support for product design. The aim of this paper is to propose a framework for an integrated software that assists designers in the early design phases to work in co-operative and collaborative manner.

Towards a Decentralized Reactive Intelligent FMS Controller

Abstract The paper presents an architecture for the flexible manufacturing control systems. The aim is to develop a particular decision-making system. The defined Controller architecture is open. It is organised into hierarchical levels according to the different decision time frames. The decision-making system is distributed among the levels, what allows reactivity and assignment of intelligence to the system. A generic model adaptable to any decision-making level, is presented. System flexibility, reactivity, robustness and reliability are expected benefits. To express system behaviour, the modelling methodology is based on the Use Case method.