Έρευνα σχεδιασμός και μοντελοποίηση ενός ολοκληρωμένου συστήματος ελέγχου και προγραμματισμού παραγωγής στη βάση ενός πρωτότυπου συστήματος αντίστροφου Προγραμματισμού Απαιτήσεων Υλικών (reverse MRP)

Abstract

This thesis discusses the design of a production planning and control system (PPC) for discrete manufacturing environments, referred to as r-MRP. The proposed approach operates under the framework of hierarchical finite capacity shop-floor modelling and discrete-event simulation. It overcomes some of the main deficiencies of MRP-based systems by considering capacity constraints, lead times fluctuations, lot sizing and priority control in a ‘bucket-less’ environment. The r-MRP system was developed based on a new method that integrates shop scheduling and material planning in a simultaneous computational process (reverse MRP). Under this methodology, material and capacity constraints are considered, together with a number of conflicting planning criteria, in a continuous time environment. Lead times are dynamically calculated variables, based on actual loads in each workcenter. A number of alternative feasible shop schedule and material plan combinations are formed in real time. The r-MRP system was implemented as an experimental software tool. Its performance has been evaluated in a case study of a textile company with ‘real life’ data. When studied through a set of simulation experiments it has been found to outperform standard MRPII system in cases of short production times, high value inventory items and frequent, small deliveries by suppliers, in terms of a number of scheduling and inventory performance indicators. The r-MRP tool can be incorporated in legacy IT systems as an interfaced functional module.