Reliable sales and demand planning
Increasing complexity and uncertain procurement markets – this is a major challenge for any supply chain. ASK Chemicals, one of the world’s largest suppliers of foundry chemicals, has coped well with this difficult market situation: The sales and demand planning process is now optimized through the use of a software tool for forecasting and compound planning. In conjunction with the ERP system, end-to-end sales and demand planning processes have been created so that increased delivery readiness can be achieved with reduced stock levels.
The ASK Chemicals Group employs 1,700 people and is a global supplier to the foundry industry. The product portfolio includes binders, coatings, feeders, filters and release agents as well as metallurgical products such as impregnated wires and master alloys for iron casting. The company sees itself as a full-range supplier. Managing such a portfolio is not easy.
Harmonized planning process
The company was formed in 2010 from a joint venture between Clariant – then still Sued-Chemie AG – and Ashland. Fabian Gasczak, Manager Supply Chain Planning at ASK Chemicals at the main plant in Wülfrath, Germany: “The heterogeneous structures for planning and controlling the supply chain must be standardized, but still take into account the individual requirements of the respective business model.” Due to the fluctuating and sometimes very short-term requirements, the processes were characterized by ad hoc planning and control actions. The high level of effort and need for coordination in the detailed planning of production was not due to a lack of planning expertise. Rather, these were only symptoms caused by the higher-level planning. Current market developments were not consistently incorporated into the planning process. These symptoms were caused by a lack of forecasts and insufficient certainty to cushion sales fluctuations. For this reason, it was decided to fundamentally restructure the planning process and support it with suitable software.
Following a comprehensive analysis, suitable methods and tools were introduced to cement the new structures that had become necessary. Since then, planning has been optimized through the use of a software tool for forecasting and network planning.
The aim of the new process is to achieve better planning results while keeping the effort involved as low as possible. Gasczak explains: “In the future, upper statistical and manual forecasts should be planned according to demand. The sales information supplements the statistical forecast in order to take current market developments into account. The sales planner then only checks the consistency of the data – which saves a lot of time and effort.” However, in order to achieve this status, the right framework conditions must also be created.
It is important to recognize that there is no rule without exception. In addition to automated scheduling, cases that cannot be forecast in a standardized way must also be taken into account. New products and product variants that cannot be forecast due to a lack of historical data must be planned manually, just like discontinued products. It must also be possible, for example, to store promotions or enter new listings and customer discontinuations.
In order to carry out precise analyses, software is also required that makes it easy to identify these exceptions. The software must therefore also be able to describe the planning at the customer’s item level and not just at the component parts list level. Mathematically and statistically, the software should also be able to automatically determine which forecasting and safety stock procedures are best suited to achieving the desired delivery readiness with the lowest possible stock levels. In practice, this selection is usually not optimal because there is a lack of know-how about the different procedures and the ERP systems used often do not allow such variant tests.
Software must be flexible
Now you might think that so many special functions do not necessarily increase planning efficiency. But that is exactly the case: analyses at ASK Chemicals showed that the proportion of automatically forecast articles is very high. For example, a number of articles are planned based on consumption using reorder point procedures – particularly in the area of raw materials. With the help of sales and forecast figures, new reorder levels are always determined and transferred to the SAP system via standard BAPls. Other articles are planned using planned independent requirements.
For these items, the pre-planning values are transferred to the SAP system with the identified strategy group. The few remaining exceptions can be planned manually by the sales department from anywhere in the world. The integrated optimization algorithms used in the simulation also result in significantly improved planning quality. The planning situations of the past are simulated and alternative methods for determining safety stocks, forecasts and scheduling as well as their parameters are applied in order to ultimately arrive at the best possible scheduling. Only a few software tools in the world are able to optimally support ERP systems in the optimization of planning and scheduling. We opted for Diskover SCO from SCT GmbH because we were most impressed by both the programming technology and the methodological expertise of this system,” explains Gasczak. The APS (Advanced Planning and Scheduling) software that we use for sales forecasting and group scheduling – developed with Java technologies – can be quickly and easily adapted to new requirements.” At the touch of a button, the methods offered can also be reduced to the possibilities of the ERP system being used. This makes the compatibility with ERP systems and the technical methodological expertise of the APS tool visible, and the ‘Sales Forecast’ module is also in use at ASK Chemicals. This module is used by the sales staff as a front end with which they can enter their forecasts at article customer level and thus incorporate them into planning. The data can be recorded worldwide and processed centrally.
Specialist article, Procurement News, Issue 11/2014By Andreas Capellmann