The Development of Virtual Product Life Cycle Design Tool Using Artificial Intelligence Technique
Keywords Virtual Manufacturing Artificial Intelligence Product Life Cycle Simulation Geometric Modeler Project Objectives 1. To design a prototype of a new virtual manufacturing processes. 2. To develop an artificial intelligence technique for product life cycle design tool. 3. To test the speed of product life cycle through virtual reality tools. Project Summary Virtual Manufacturing (VM) is the use of computer models and to aid the design and production of manufacturing products [1,2]. VM was first used in 1982 by McLean et. al. [3] when they were looking at ways to extend Group Technology (GT). This term has also been used more recently by Mills et al.[4] who investigated software for designing and simulating manufacturing cells for products designed on CAD packages or geometric modelers, and by Kimura [5], who has coupled the use of knowledge-based design with the modeling of engineering objects and the simulation of manufacturing on computers. VM implies interaction with a manufacturing environment as achieved by Barrus[6] with the creation of a Virtual Workshop containing a selection of manually operated machines. The system uses the ACIS[7] geometric modeler, exploiting the internal roll-back mechanism to perform multiple undo and redo of operations. VM is expected provide accurate estimates for processing times, cycle times and cost, as well as product quality. Three AI techniques are fuzzy logic (FL), genetic algorithm (GA) and neural network (NN). The application of AI can be found in many areas such as in comsumer clasification [8], diagnosis system[9], longevity modeling[10], and vision tracking [11]. Therefore, the application of AI technique in development of virtual product life cycle design tool should be considered. The combination of NN, GA and FL might be considered based on the work[12] that combine GA and FL. The purpose of this project is to apply AI technique namely using one of these three techniques or hybrid of them. Works done [13,14,15,16, is preliminary work done that contribute to this proposal. Research methodology The development of virtual product life cycle design consists of the following steps: (i) Product Model Definition, (ii) Device Model Preparation, (iii) Virtual Product Design, and (iv) Simulation of Manufacturing Processes. Those steps are explained briefly in below. 4.1 Product Model Definition In this stage, product data used for fabricating products such as product model, cost estimate, and so forth will be investigated. Thus, potential problems with the design can be identified and its merit (in forms of cost) can be estimated. 4.2 Device Model Preparation Device model database will be developed according to the real device specification that involved in developing actual product. This is intended to calculate the behaviors of devices. Object-oriented
modeling technology will be used for representing device’s behaviors in VM system. Object-oriented technology provides a rich way to classify the devices and their relationships in manufacturing systems. By using this approach, the characteristics of manufacturing systems can be identified and described by classes of objects and their relationship. 4.3 Virtual Product Design AI based design methodology of embedded controller will be used to reduce the design development cycle Sequence of design steps required to develop a controller using Conventional and a AI approach are compared. Steps involved in conventional approach are: (i) Understand physical system and its control requirements, (ii) Develop a model which includes the plant, sensors and actuators, (iii) Use linear control theory to determine a simplified version of the controllers, (iv) Develop an algorithm for the simplified controller, (v) Simulate the design including the effects of non-linearity, noise, and parameter variations. If the performance is not satisfactory, we need to modify our system modeling, re-design the controller, re-write the algorithm and re-try. On the contrary, few steps are involved in AI based design methodology. Those steps are: (i) Understand and characterize the system behavior, (ii) Design the control algorithm using AI rules, which describe the principles of the controller’s regulation in terms of the relationship between its inputs and outputs, (iii) Simulate and debug the design. If the performance is not satisfactory, we only need to modify some AI rules and retry. 4.4 Simulation of Manufacturing Processes Simulation is an invaluable tool in designing and analyzing manufacturing systems. Simulation of manufacturing processes will be carried out using ACIS. ACIS is a software interface to graphics hardware that allows a programmer to specify the objects and operations involved in producing highquality graphical images. ACIS is a set of commands that allow the specification of geometric objects in two or three dimensions, together with commands that control how these objects are rendered into the frame buffer.