A genetic-based approach to simulation of self-organizing assembly

Abstract

The paper proposes a new and innovative biologically oriented idea in conceiving intelligent systems in modern factories of the future. The intelligent system is treated as an autonomous organization structure efficiently adapting itself to the dynamic changes in the production environment and the environment in a wider sense. Simulation of self-organizing assembly of mechanical parts (basic components) into the product is presented as an example of the intelligent system. The genetic programming method is used. The genetic-based assembly takes place on the basis of the genetic content in the basic components and the influence of the environment. The evolution of solutions happens in a distributed way, nondeterministically, bottom-up, and in a self-organizing manner. The paper is also a contribution to the international research and development program intelligent manufacturing systems, which is one of the biggest projects ever introduced.

Introduction

Production of goods is surely one of the main human activities. The goods are of various types and can be material ones (cars, computers) or non-material (software). Any of the goods is, in fact, a more or less complicated system which must be produced with the material, energy and information flows synchronized as much as possible.

Nowadays we face the increase of the number of versions of goods and shortening of their service life. The individualization of man's desires and striving for ever newer goods is the generator of such trends. The individualization of man's desires has in fact been existing already for a long time, but it has not expanded because the products could not have been produced within acceptable frames of time and money for lack of sufficient technical support (computers, software, machines). Due to the flood of highly capable computer programs and advancing production technology now the possibility is given to make modifications of products quick and more affordable. The entire progress of production of a product, from the concept to the manufacture, can be simulated in the virtual world of suitable computer programs and only when the product has successfully undergone all simulations, the actual manufacture starts. In this way a significant portion of the total costs, likely to occur if manufacture was carried out in the real world at the very beginning, could be avoided.

However, the evergrowing number of products and shortening of their service life do not have only positive effects. The loading of the environment and consumption of raw materials for the manufacture of goods are increasing sharply. Man is a being that consumes about 100 times more energy per unit of body weight than any other living creature. That inevitably leads to exhaustion of the natural resources of the planet. The modern methods of production of goods are becoming increasingly energy saving, but the number of new products and the consumption of energy increase faster than the savings of the energy-saving production. In addition, in the production of goods the emphasis is put particularly on the mechanistic-technical aspect emphasizing in particular the technical advantages of the goods and almost not at all taking into account its interactions with the wider environment and the ecosystem. Optimization of manufacture of products takes place merely from the technical point of view and is limited only to carefully selected parameters of the optimization — only those which we know and only those which we want to know [1]. The optimization is thus limited only to a small subsystem of the wider system. It is clear that in this deficient way it is not possible to reach the global optimum of production, in particular not if the wider environment is considered. For example, the modern production is almost possessed with the desire to produce as many products as possible in the shortest possible time. If the products are actually sold, the optimum has been reached. But it is questionable if the analysis of the product properties includes enough influential parameters, so that this optimum is global (e.g., in the technological sense the product may be special, but it can be harmful to the health of man and to the ecosystem).

The existing production concepts including the so-called flexible manufacturing systems cannot successfully respond to the above-mentioned challenges to which the modern production will be exposed in the third millenium [2]. Inflexibility of the present production systems is the cause of many difficulties expressed particularly as too slow adaptation of production means to new requirements, which results in high costs of the product and bad adaptation to the environment in which they exist. Therefore, at the transition into the new millenium new forms and concepts of production systems are searched for. The new production concepts try to combat efficiently with the technical, economical, and ecological requirements of the modern time. In the literature it is possible to trace at least three modern concepts of manufacturing systems: fractal, holonic, and bionic manufacturing system concepts [3], [4]. So far, none of these concepts has been more widely introduced into practice; however, enormous funds and efforts have been invested to ensure that the manufacture of goods in the future would be more intelligent, more efficient, cheaper, and adapted to the ecosystem. The international intelligent manufacturing systems (IMS) program is one of the projects having just this objective [4].

In the future we will face the increase of today's trends towards individualization of products and shortening of their service life, therefore intelligent production decisions are inevitable [5], [6], [7], [8]. The entire production tends towards small series, with many possible scenarios, where intelligent decisions have a greater advantage over the ‘hard’ automation and flexible manufacturing system.

The objective of the presented research is to make a contribution to the endeavours to ensure that production of goods in the future is as intelligent as possible. First, in short the description of the genetic programming method is given. Then we study the properties of the living and artificial systems. Some properties of the living systems are then adapted to the elementary parts which are the basic components from which the product is formed. The concept of the self-organizing assembly of the elementary parts into the final product is given in short. Then we perform the simulation of the self-organizing assembly of the product on the basis of genetic content in the basic components and the influence of the production environment and the environment in a wider sense. The paper ends with an analysis of the work.