Scheduling (production processes)

As sequencing or sequence planning ( in the scientific field, even under the English term sequencing and scheduling familiar ) is referred to in production planning the formation of a production sequence of production orders. In many industries (chemical industry, automotive industry) it is inconvenient to execute production orders in the order they arrive. The production sequence is instead designed so that, for example, the utilization of equipment and manpower evenly or set-up costs of machines are minimal. Produced in a company producing batches, not individual production orders, but all lots are sequenced.

Is sequenced due to the usually high Rechenaufwändigkeit often only a part of all this production orders. Which may be, for example, the range of products of a single layer or one day. The assignment of orders to day or shift programs is referred to in the literature as mixed - level scheduling, smoothing or balancing. This then results in planning hierarchies such as Month - Day - clock.

In American literature, the terms scheduling and sequencing (English Sequencing ) are often equated. However, it should be noted that a sequence is usually an immutable fixed production sequence designated (see also [ pearl necklace ] ), while the scheduling generally describes the production sections in chronological order in which the location of a job or multiple jobs in retrospect may change. (See also scheduling. )

Scientific approaches and methods to solve the sequencing problem

To meet the requirements for a technical solution to the problem for determining an optimized production sequences meet the experts of different approaches to use. Basically two principles are applied in this case.

Calculation of a mathematically exact solution determined using:

• Constraint programming
• Mathematical programming ( integer programming )
• Ad - hoc method
• Branch and Bound Method ( german).

Finding a solution using heuristics such as:

• Greedy algorithms ( greedy approach)
• Local search ( Local Search Approach)
• Evolutionary Algorithms (Evolutionary Algorithms )
• Ant algorithm ( Ant Colony Optimization).

Combinations of the above methods can be applied.

Systems for solving sequencing problems

Stark fallen investment costs for computer systems already enable the use of modern, high-performance software solutions that build on the basis of the order data from job leading ERP / PPS systems together with the specific requirements of the production. Most of these tools make use of this so-called solvers that are based on the methods already described above and, for example as a function libraries are available on the market. Extended with additional features, integration into existing PPS / MES environments is simplified and in the first place.

Increasing complexity and increasing variety of products to be produced meant that the automotive industry when it comes to the use of sequencing systems, has taken a leading role. Other industries which have meanwhile become oriented to the production principles of automobile manufacturers [ see also Toyota Production System] think, increasingly request on the use of such planning tools. In particular, because such calculations not only to serve the determination of actually producing days needs, but also because it is possible to simulate different scenarios for analysis and the results of simulation runs to compare.

In addition to the equipment and labor capacity already mentioned can be considered as restrictions other assembly -related aspects such as:

• Different cycle times in heterogeneous production lines
• Distances to be observed
• To be formed blocks (eg summary of same-colored bodies in the painting of vehicles )
• Wedge-shaped restrictions (eg at the beginning of a layer only a few complex vehicles)
• Equal distribution of time-consuming work content.