Tibor Tóth

On Scheduling Approaches of Flexible Manufacturing Systems: Gap Between Theory and Practice

Abstract Due to the advances and developments in both hardware and software, the technology embedded in Flexible Manufacturing Systems (FMSs) is not only becoming more complex, but also presenting a number of decision problems to be solved, especially in the area of scheduling. An exhaustive survey of the literature on FMS scheduling clearifies several significant differences between what the theory provides and how it is applied in practice. The main objective of this paper is to highlight the gap between the theory and the practice in the field of FMS scheduling. In this paper, scheduling domain is decomposed in a top-down manner, and an interactive scheduling system, based on the information and data flow on this decomposition, is proposed to bridge the gap.

CIM: ON A NEW THEORETICAL APPROACH OF INTEGRATION

Abstract It is expedient to study up-to-date theoretical means and practical resources of the automation of mechanical engineering through the structure, functions, environment interrelations and the most important features of Computer Integrated Manufacturing (CIM) systems because the highest level of dual unity of automation for material- and data processing can be realized by them. Classification of CIM systems and their subsystems is generally based on the automation levels and functional roles, i.e. on the external features of the systems and subsystems. In this paper the authors will prove that there is another classification method well-established from the theoretical point of view, which determines the internal hierarchy of part manufacturing systems on the basis of objective features. Starting from a uniformable model of these systems the process optimization principles of discrete manufacturing will also be generalized.

Intensity type state variables in the integration of planning and controlling manufacturing processes

Abstract Nowadays rapid changes of market conditions require more efficient integration of Process Planning and Production Controlling in manufacturing industry. In computer integrated systems the classical models of manufacturing operation planning arenot adequate for flexible operatio management. This paper shows a new conception to use intensity type machining state variables (e.g Q [ cm 3 / min ]) which expresses both machining and management aspects at the same time. The theoretical background of the conceptual and some application examples will also be detailed.

The “production triangle” model in Production Planning and Control

Capability to adapt to the changing requirements of customers presents an increasing importance in the field of industrial production. Increasing demands of quality and reliability can only be met with a high level of readiness for delivery. The up-to-date ERP systems include such Production Planning and Control (PPC) components that are based on theoretically well-established state-equations, production models, data-base transactions and functional supplies. Three macro-parameters: readiness for delivery, stock level and utilization of the available production resources can form such an abstract model — the “production triangle” — on the basis of which a model meeting both the necessary and, at the same time, the satisfactory conditions to realize a good approximation of optimum production planning and control can be created. One of the essential problems of PPC is that the readiness for delivery, stock level and utilization of production resources are such complex state variables (macro-parameters), which cannot be managed independently of each other, especially when disturbances and/or uncertainties occur in production processes, which cannot be planned before executing the processes themselves. The utilizable application systems have to present hierarchical structure along time domain. Long, medium and short term production planning are to be combined with real-time control of manufacturing processes. In this sense, the Manufacturing Execution Systems (MES) may offer new possibilities. Solving of hierarchical production planning and control tasks is fairly difficult, despite of the advantageous achievements in the field of Information and Communication Technology (ICT). The objectives defined for a higher hierarchy level are transformed to the lower levels in form of constraints. In the course of the last few years the authors have got advantageous experience related to utilization of “Production Triangle” based approach, both in mathematical modeling of the objectives and constraints, as well as in elaborating the methods and algorithms suitable for constituting the basis of new algorithms and software prototypes.

On Some Functions of the MES Applications Supporting Production Operations Management

Integrated computer applications play an increasingly important role in the planning and control of production systems and processes. The model-based decision support functions of business and manufacturing processes can be classified into different hierarchical levels according to their functions, objects and time horizons. Manufacturing Execution System (MES) applications represent the most challenging part of planning and control activities in Manufacturing Operations Management (MOM) in supporting its Production Operations Management (POM) activities. The ISA-95 standard delivers useful methodology and several models with the most important functions to be embedded in the shop floor level control loops. Two methods for improving the quality of POM are presented: a proactive one, using simulation-based fine scheduling and a reactive one, based on evaluation of some Key Performance Indicators (KPIs) determined from the efficiency analysis of shop-floor production data. Both methods exploit advantages of software applications used in different MES components.

Interactive Programme System for Determining the Optimum Machining Tolerances Having Regard to Assembly Requirements

The paper presents a mathematical model and a solving algorithm by means of which, with knowledge of the tolerance prescribed for the resultant of an assembly dimensional chain containing components of optional number, the optimum tolerance ranges of the free-toleranced components can be determined. The minimum of machining costs is considered as an optimum criterion. The dimensional chain may be of linear or non-linear type. Total or limited changeability can be selected as an optional supplementary condition in accordance with the user’s wish. For numerical solving the mathematical model, a PC AT based interactive programme system was developed by means of which economical tolerances of the free-toleranced components can be determined in three optional ways: making use of the optimization model and solving algorithm based upon Lagrange-multi-plicators; with the aid of the equal distribution of tolerances; using the principle of the equivalent degree of tolerances.

Optimisation of Discrete Technology Processes Using A Method Traced Back to Constrained Travelling Salesman Problem

In computer aided planning of discrete technology processes the modern methods of Operations Research have an important role. Application of computers also enable to enforce optimization strategies in planning of these processes.

Multi-objective performance analysis of one-machine manufacturing systems

Achieving of the desired performance for manufacturing systems is one very important goal when solving of planning and scheduling tasks. Corresponding management of orders and invested assets demands for multi-level and multi-objective optimization. Software systems of Enterprise Resource Planning (ERP) and Manufacturing Execution System (MES) applications used presently in practice are based on different modeling approaches. However, most of involved mathematical models use a set of complicated production equations. In these equations, released jobs and operation rates (as input) can be considered as the most important parameters that may play the controlling role of the production. The considered mathematical models usually evaluate three key performance indices (KPIs) of great importance. These indices are as follows: 1. Readiness for delivery, 2. Stock or WIP (work-in-progress) level, 3.Utilization rate of resources. These indices can be considered as objective functions not only for manufacturing systems, but for the entire enterprise, too. Unfortunately these indices depend on each other and influence each other in an opposite way. A basic requirement of control policy is production stability. In the case of stable production, there is a general correspondence among the average performance indices together with the average released job set or production-rate parameters. These are analyzed for one-machine manufacturing systems.

Computerized Determination of Optimum Cutting Conditions Having Regard to the Permissible Tolerances

Abstract Determination of optimum cutting conditions is one of the important and systematically returning tasks of Computer Aided Production Engineering (CAPE). In the case of turning slim stepped shafts with high accuracy, the permissible values of cutting parametres depend on the resulting elastic deformation of the current machining system to a great extent. The paper deals with the theoretical basis of computing the extremum values both of resulting elastic deformation of a given machining system for turning and of diameter errors for all the surfaces to be machined. A PC AT/386 based program package suitable for solving these tasks will also be outlined.

AI and machine learning research within the framework of a CIM pilot system

Abstract This paper reviews the role of CIM in the education of a new generation of mechanical engineers. Accepting the increasing importance of computer aided engineering techniques, the structure and modules of a CIM Pilot System under development will be outlined. At present there are already some important fields of CIM where AI methods and Machine Learning can be considered as indispensable tools for development. A function scheme suitable for Machine Learning in the FMS/CIM Pilot System of the University of Miskolc will be described in detail.