Customers expect timely deliveries, factories with good production planning run more efficiently. These are the external and internal factors that enable precise OEE analysis and use of data. How must production planning and OEE analysis work together to achieve these results? We provide answers.
Production planning only makes sense if it provides a minimum of precision. The APS/PPS tools available on the market all have the classic “garbage-in garbage-out” problem of information technology: They can only plan as well as they are supplied with data.
Two major and generally unused levers for increasing planning quality are fairly consistently noticed in factories: Incorrect expected values for plant occupancy time with a production order and incorrect assumptions of the set-up time between two orders. This results in unstable planning, large parts of which have to be rescheduled again during the next planning run. No sign of good planning. How to prevent this phenomenon?
Precise calculation of system occupancy time
In order to meet the planning problems of manufacturing on the shop floor, the plant capacity available according to shift models is often adjusted downwards by a flat factor. The system is therefore no longer available with 100% capacity, but only with 78%, for example. The value was calculated using a long-term average. If the planning performance is not achieved several times in a row, the value is adjusted further downwards. This discounting is intended to take account of the many minor disruptions that occur during order processing.
However, reality is more dynamic and can be represented more precisely. The adjusted difference is in fact OEE (minus losses due to set-up). With oee.ai, this can be identified per system and per product and passed on digitally and in real time to an APS/PPS system. The calculation horizon in the past can be freely selected.
In this way, you not only do justice to product differences, but can also incorporate improvement measures in the production process into planning without manual parameter adjustment.
Precise planning of set-up processes
Larger planned downtimes such as maintenance or cleaning shifts are usually reflected in the planning and the times are also met on the shop floor, so that these rarely lead to planning deviations. However, the situation is different with set-up times. We see companies that upgrade their plant pool 40,000 times a year, but plan this with a lump sum of 30 minutes.
Here, too, more precision is possible by using oee.ai. Modern OEE systems automatically generate a setup matrix in the background, i.e. show the duration of the set-up process for all combinations from previous product to next product. These real times can be assigned a statistical variance and also automatically transferred to the APS/PPS tool so that they can be incorporated into the planning there.
What is left of the lack of precision?
Two other factors that cause frequent turbulence in factory processes have not yet been addressed. Short-term and unexpected reduced employee availability due to sick reports at the start of the shift cannot be taken into account in planning in this way, nor can unexpected prolonged technical faults be taken into account. In terms of planning, these are individual events for which no statistical basis is readily available.
Feel free to contact us if you have any questions about the integration of PPS/APS systems and oee.ai.