Tonči Mikac

Evolutionary approach for solving cell-formation problem in cell manufacturing

New environmental characteristics and requirements have challenged manufacturing enterprises to improve the efficiency and productivity of their production activities. Cellular manufacturing (CM) as one of the solutions to this problem, as implementation of Group technology (GT), was proposed. Performance of cellular manufacturing systems depends heavily on the cell structure. In this paper we present and evaluate a novel approach for solving the cell-formation problem based on implementation of a GA, called modified genetic algorithm (MGA). Implementation of the GA in here presented simulation model is based on Emergent Synthesis idea, of top-down problem decomposition and emerging solution from bottom-up feedback.

Establishing an Institutional Framework for an E-learning Implementation – Experiences from the University of Rijeka, Croatia

Faced with the need of transforming the university structure, processes, and programs according to the Bologna reform, and in order to become more fle

A STUDY OF FACTORS THAT AFFECT THE BUILD-UP MATERIAL FORMATION

A feature when cutting many alloys is that workpiece material adheres to the cutting tool at the sliding contact surfaces, between the work material and the tool. This built-up material formed during cutting is of fundamental importance in machining operations, because it may significantly affect the surface roughness, tool wear, workpiece dimensions and tolerances, tool forces, and chip form. The agglomeration of the work material to the tool appears to be analogous to cold welding, metal transfer in tribology and dead zone in extrusion. In machining terminology this phenomenon is often called “built-up edge” (BUE). Several important factors affect the built-up material formation, e.g. cutting temperature, cutting speed, strain hardening, adhesion between the work material and the tool, micro-crack formation, plastic flow of the work material in the vicinity of the cutting edge, etc.

A NEURAL NETWORK APPROACH IN CELLULAR MANUFACTURING

In order to achieve high productivity under the turbulent production environment, production systems have been changed into cellular manufacturing systems. That involves dividing of production system into some cells by collecting similar products and machines in manufacturing process. In this paper, a neural network approach for solving cell formation problem is investigated. We modify the binary self-organizing neural network algorithm (ART1) by changing the weight vector updating equation and using a set of supplementary procedures. The effectiveness of the proposed algorithm is shown by comparing of grouping efficiency with some other well-known approaches like rank order clustering method and ARTlneural network (original and another with reversed ones and zeros).

Multi-Agent Manufacturing System Design Based on Emergent Synthesis Approach

This paper presents a approach for multi-objective design and reconfiguration of the current manufacturing system. A proposed model for multi-objective design and adaptable reconfiguration is based on the multi-agent principles and Emergent Synthesis concept. Emergent Synthesis is introduced in the adaptable reconfiguration within implementation of the layered based architecture and utilizing the multi-agent concept. Based on the proposed approach a simulation model has been built and evaluated. For the purpose of analysis of this approach it was simulated production in the shop-floor of the manufacturing system with the two different initial configurations. For evaluation of here presented approach the case study was performed using realistic data; it was simulated production of the cellular phone devices. From obtained results of the simulation it can be conclude that a proposed model has a high level of multi-objective adaptation of the manufacturing system shop-floor.

A MOBILE ROBOT PLATFORM FOR FERROMAGNETIC BASE PLATES

In this paper a prototype of a robot platform designed for movement on large ferromagnetic plates is presented. The intention was to build a platform with a high positioning accuracy, which have an unrestricted working area and carry a universal robot on its top plate. Throughout this paper the main elements of the mobile system is presented. The robot platform has a hydraulic drive motor, 7 cylinders and six electromagnetic legs. The movement of robot can be linear or rotational in both directions, which enable it to reach any position on an extended surface.