MasDISPO_xt: heat and sequence optimisation based on simulated trading inside the supply chain of steel production

The production of steel normally constitutes the inception of most Supply Chains in different areas. Steel manufacturing companies are strongly affected by bull whip effects. Due to nondeterministic incoming orders and changes of customer requirements on accepted orders, making the right decision at a certain stage can be the difference between earning or loosing a great turnover. Improving their operational efficiency is required to keep a competitive position on the market. Therefore, flexible planning and scheduling systems are needed to support these processes which are based on considerable amounts of data which can hardly be processed manually. Existing systems are dominated by centralized decision making processes, mostly data driven and often not modeling the business processes they should. MasDISPO xt is an agent-based generic online planning and online scheduling system for monitoring of the complete Supply Chain of Saarstahl AG, a globally respected steel manufacturer. This paper concentrates on the creation and optimisation of heats and sequences as a presetting for the production inside the steelwork. 1. PRODUCTION AT SAARSTAHL AG The production chain of the Saarstahl AG consists of a multitude of specialized and complex metallurgical manufacturing processes with a lot of interdependencies among them. First, a blast furnace factory produces hot metal for the steel production. In certain intervals equally distributed over the day, a certain quantity of hot metal is sent by rail to the steel works for the next production step. Inside the melting shop, steel of different quality grades is produced according to concrete customer orders and requirements. It is cast at continuous casting plants into billets. A single production unit inside the steel works is called heat grouped together in sequences – a totallly ordered set of similar heats of equal formats and qualities. Cite as: MasDISPO_xt: Heat and Sequence Optimisation based on Simulated Trading inside the Supply Chain of Steel Production, S. Jacobi, D. Raber and K. Fischer, Proc. of 7th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2008)Industry and Applications Track, Berger, Burg, Nishiyama (eds.), May, 12-16., 2008, Estoril, Portugal, pp. 23-26. Copyright c © 2008, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved. Figure 1: supply chain of Saarstahl AG Afterwards, these billets are delivered to the rolling mills. Here, steel bars and wire rods of different shapes and formats are produced. In fixed, cyclic rolling campaigns of limited capacities certain formats are produced. These cycles are dependent on the rolling mills, billet supply of the steel works and explicit orders from customers. They vary between one to four weeks. After the rolling, potentially following processes concerning steel bars are arrangement, pickling, annealing and saw cutting; wire rods probably need a annealing, a pickling or both. Finally, the products are delivered to the customers – mostly suppliers of the automotive, shipbuilding or aerospace sectors. Figure 1 depicts the roughly described Supply Chain. Given a working plan, the system schedules the execution of each concrete order along the production chain. It monitors production on a rough—in weeks—and detailed—in days and hours—level, and executes an online detailed planning and scheduling for the different manufacturing phases. It has to detect problems in the production and handle them in order to return to normal production. The rough working plan for each manufacturing phase is calculated on demand, before final order commitment. Depending on delivery date, order size and vertical integration certain capacities at specified aggregates have to be roughly allocated.