ERK'2014, Portorož, A:159-162 159 Automated simulation model building for a complex furniture manufacturing process Tadej Kanduč 1 , Blaž Rodič 1 , 1 Faculty of information studies, Ulica talcev 3, Novo mesto, Slovenia E-mail: tadej.kanduc@fis.unm.si Abstract In this paper we present the first part of a project of flexible manufacturing process optimisation in a Slovenian furniture company. In this part we analysed the current state of the manufacturing system by constructing a discrete event simulation model that reflects the manufacturing processes. As the manufacturing process involves about 30,000 subprocesses it is not feasible to manually construct the model to include every subprocess, hence a method for automated model construction was developed. The method analyses open orders and constructs a corresponding model which includes only the machines and subprocesses required for the manufacture of ordered products. The obtained simulation model lets us inspect current manufacturing processes and allows us to easily apply modifications of the model parameters in manufacturing process optimisation. 1 Introduction Optimising manufacturing processes is one of the key goals in every manufacturing company. If the processes are simple enough, they can be efficiently represented by exact mathematical model which can be effectively optimised with analytical mathematical algorithms. In practice, the processes are usually too complex to tackle them with this approach and a common alternative is to construct a discrete event simulation model (DES) in order to understand and analyse every stage process in a floor. Our primary focus in the project is to investigate how the layout of machines on the factory floor affects the efficiency of manufacturing processes and to optimise the layout. We restricted our criteria of efficiency to measuring total time it takes to complete an order, machine utilisation, and total distance the manufactured products need to make on the floor. In this paper we describe the method of automated model building used in the first part of the project. 1.1 Description and complexity of manufacturing processes The furniture company uses approximately 140 machines and produces over 30,000 different products. To manufacture a final product, each product has its own technical procedure, i.e. a sequence of prescribed tasks. For each product and task there is a list of suitable machines and expected completion times. More complicated products are manufactured by joining smaller semi-finished products according to the prescribed bill of materials (BOM). After a task at a machine is finished, the entire series of products is moved by carts to the next prescribed location (machine). List of all final products that need to be manufactured is determined by the clients' orders. Due to the large set of different products and dynamic nature of the manufacturing processes, static model development is not feasible in this case. Instead, the model is built automatically, based on available machines and for current open orders. Technical procedures and BOMs are read dynamically from input data during the simulation. 1.2 Previous research (review of literature) Simulation is commonly used for the evaluation of scenarios [1],Napaka! Vira sklicevanja ni bilo mogoče najti.,Napaka! Vira sklicevanja ni bilo mogoče najti.. However, the models developed with the visual interactive modeling method (VIM) are usually manually constructed through careful analysis of the real-life system and communication with process owners. Automated model development is more common with methods that allow easier and more standardized formal description of models, e.g. Petri nets [4],[5]. Automation of model construction and adaptation can importantly facilitate the development of models of complex systems Napaka! Vira sklicevanja ni bilo mogoče najti.,[7] and generation of simulation scenarios. Several papers deal with factory layout optimisation, with the paper Napaka! Vira sklicevanja ni bilo mogoče najti. stating that multiproduct enterprises requires a new generation of factory layouts that are flexible, modular, and easy to reconfigure. Evolutionary optimisation methods are often proposed due to optimisation problem complexity Napaka! Vira sklicevanja ni bilo mogoče najti.. Factory layout design optimisation is further discussed in Napaka! Vira sklicevanja ni bilo mogoče najti.,Napaka! Vira sklicevanja ni bilo mogoče najti.,Napaka! Vira sklicevanja ni bilo mogoče najti. . Authors Napaka! Vira sklicevanja ni bilo mogoče