The present paper aims to develop a simulation tool for tile manufacturing companies. The paper shows how simulation approach can be useful to support management decisions related to production scheduling and investment planning. Particularly the aim is to demonstrate the importance of an information system in tile firms. The Factory Data Model (FDM) parameter is used to describe the activities in ceramic tile industries operating in different European countries. A process-based analysis of tile manufacturers is undertaken and individual company performance is quantified by Key Performance Indicators (KPI). The overall model is composed of different processes, which are coded into Scilab environment and matched together to arrange a stochastic simulator. The simulations results are used to show how management decisions can significantly effect the KPIs. The simulations highlight the effects on KPIs of three specific parameters: the length of scheduling period, the quantity of stock needed and the reliability of the information system supporting orders. The results clearly show that the effect of allowing the presence of unattended orders within the outstanding orders list always has a remarkable negative influence on KPIs. Results also suggest that the presence of sub-groups of homogeneous tiles, based on colour variation, is one of the most important factors affecting a tile manufacturer's performance. The results of the simulations have two different practical implications. Firstly, they demonstrate the importance of information systems in tile companies, suggest to evaluate investment in information technology and indicate the value of promoting an information culture in the entire work forces. Secondly, they show the potential of simulation tools development to support decision making in a BPR (Business Processes Re-engineering) scenario.

No abstract

Abstract. The economic order quantity (EOQ) and the economic production quantity (EPQ) are well-known and commonly used inventory control techniques. The standard results are easy to apply but are based on a number of unrealistic assumptions. One of the assumption is that the demand is normally distributed in any interval. In several practical cases the assumption about independence of successive demands, and consequently demand normal distribution in any interval, is not supported by real data. This paper investigates the effects on the expected service level (SL) after relaxing normal distribution assumption on the demand. The present work shows a possible strategy to use classic inventory model, such as EOQ/EPQ model, adopting discrete event simulation analysis to quantify model performances under relaxed assumptions.Keywords: EOQ, inventory techniques, discrete event, simulation analysis. IntroductionThe economic order quantity (EOQ), first introduced . The standard EOQ and economic production quantity (EPQ) results are easy to apply but are based on a number of unrealistic assumptions [6]. One of the assumption is that the demand is normally distributed in any interval, it is assumed that successive demands are independent and, consequently, the accumulated demand over many time units is approximately normal. The realization that inventories operate under less than ideal situations gives rise to a subset of inventory modeling theory that performs sensitivity analysis on models operating under stochastic conditions [7]. Several extensions of the classic EOQ/EPQ model have been, Borgonovo [8] presents a good review of them across several fields of research. A branch comprises models where the assumption that all units are of perfect quality is removed, for a deep literature review you can see Chan [9]. Another field of EOQ extended models is focused on deteriorating inventory models for perishables management, for a good review you can see Goyal [10] and Ferguson [11]. Some papers mixed the two proposals to develop a model that better

This paper investigates the re-order point inventory management models sensitivity to demand distributions, demand dependencies and lead time distributions. The investigated performance measures are four different versions of service level. The conclusion is for all measures that the single most critical aspect adversely affecting service level performance is the presence of asymmetrically distributed demand.

No abstract