DEUTA Werke GmbH

Background

DEUTA Werke produces light engineering and electronic sub-assemblies and equipment for fitting out rail vehicles at its Bergisch Gladbach site. The products include, e.g. sensors for mechanical variables, indicators and complete driver's cab indicators. The production is divided into four manufacturing cells, which produce the sensors, indicators and electronic sub-assemblies and assemble and integrate these components. The production is controlled by process-oriented methods. The individual workstations are specifically set up to produce and/or test the products according to the customers' wishes. The employees are capable of working on different products in several of these areas.

Task

The declared objective of DEUTA is, as far as possible, to meet all the deadlines agreed with customers. The prototype planning software to be developed within the scope of Signal should initially, above all, help to achieve this goal. Because the company did not agree to participate in Signal until May 2000, there was a very small time budget available, which had to be optimally used. Therefore, the knowledge acquired at Albert Ackermann GmbH & Co. KG was taken into account, in consultation with DEUTA, in an action plan which described clearly defined and manageable sub-steps. The initial priority was to produce a validated model suitable for deadline & scheduling forecasts.

Implementation

The structure of the simulation model shows the processes in schematic form (Fig. 1). The simulation model has a hierarchical structure, so that the processes can be "opened" by clicking them and the large number of model elements (workstations, buffers, control methods) becomes visible. This is exemplarily shown in (Fig. 2). The simulation was stopped and then showed the momentary location of the orders, symbolically displayed as a "crate".

deuta1Figure 1: Schematic display of the processes at DEUTA

deuta2Figure 2: Detailed display of a process

DEUTA has a PPS system, which contains a large amount of the data required for the simulation. The company quickly realised an export facility for order data, duty rosters and production plans. In addition to the PPS data, the simulation also takes into account the employees' qualifications. This data is managed by the planning system, because it is not available in the PPS.

Results

The features of the SIGNAL prototype made available for the trial operations extended far beyond the planning task. It was possible to examine the employee times and qualifications and a change in employees between the individual manufacturing cells and their effect on the weekly result. Apart from the pure feasibility, the results of the simulation show the capacity utilisation of the individual resources and the schedule effectiveness and meeting of deadlines for the individual orders. The employees involved in the trial run found good correspondence between the order data of the existing PPS system and the simulation model. The main difference between the simulation and the PPS system was identified as being the detailing of the employee consideration with respect to the working hours of the individual employees and their qualifications. Various versions of the weekly plans were tested to compare the different solutions with each other. In particular, it was possible to plan the employee times.

Conclusion

A planning tool was made available to the group leaders of the various processes with which they were able to simulate their weekly plan in advance and to optimise it by manually varying the parameters.