Disposition strategies

If suppliers to DIY stores want to offer their customers full shelves and good margins at all times, it is not enough to feed the JIT Supply Chain chain 4.0 from the finished goods warehouse or the last available machine.

DIY stores always want full shelves – ideally with fast-moving items and good margins. Suppliers sometimes provide their own rack jobbers or hire sales service companies to present a perfectly stocked range at all times. With solutions such as “Logistics 4.0”, the aim is to bring sales figures even closer to the point and ideally respond with a batch size of one. However, all these efforts do not lead to success if suppliers always want to meet these sometimes highly fluctuating demand reports as quickly as possible, for example within 24 hours.

To do this, they would have to maintain very high stocks in the finished goods warehouse or JIT capacities for production. Both are superfluous outside of peak times and are therefore expensive and tie up capital. This is why supplying companies are looking for ways to implement all of this more cheaply in order to ultimately be able to offer attractive prices to the construction market. The potential is enormous: in addition to tying up capital, which leads to dead capital, there are as many as 18 to 30 other costs in the inventories resulting from capital costs, insurance, administration, storage capacity, etc. These costs must be paid by the DIY store. The DIY store has to pay these costs if the logistics chain is not right. But how can you actually increase delivery readiness and reduce stocks at the same time?

Optimize scheduling processes

First and foremost, it is a question of better scheduling processes. For example, deliveries can be made more quickly and at shorter intervals. This reduces storage capacity. In turn, products that are rarely in demand can be produced on demand and removed directly from the finished goods warehouse. In addition, the logistical decoupling point can be set as far as possible towards the end of the supply chain through modularization, thus reducing inventories across the entire supply chain. Many logistical variables are also planned by hand and executed manually. Filling a pallet space in the truck with slow-moving items just to save freight costs quickly drives up stock levels. It is therefore important to optimize many things across the entire supply chain.

One company from the metal industry that supplies DIY stores, for example, has managed to significantly reduce its stock levels while increasing its delivery readiness: GAH Alberts. The manufacturer of fittings, profiles and fencing technology, among other things, was able to reduce stocks by 13% in the short term and a whopping 53% in nine months. More than half of the finished goods warehouse was therefore stocked with material that was not immediately required in order to maintain a high level of delivery readiness.

Another supplier for DIY stores in the lighting sector had to overcome the challenge that the components for the products, which are mainly manufactured according to the company’s own designs, were largely procured in China. The delivery times here are between 60 and 150 days. However, the customers, specialist and wholesale companies, demand the highest delivery readiness at all times. Around 20,000 customer order items are received every day, most of them with a delivery time of 24 hours. A sufficiently high level of stock is therefore required on the procurement side. However, if the demand situation changes on the customer side, items that were previously procured well in advance are no longer needed and become excess stock. Such problems therefore need to be managed.

Tool skills

Methods and tool skills are required for this. At the lamp manufacturer, for example, an extended ABC analysis was carried out. In other words, a classification of the complete product range according to

• economic importance,
• Regularity of consumption,
• Number of customers per item and
• Life cycle

These classification characteristics are important parameters for deciding which planning and scheduling parameters should be set for which article. In addition, a set of rules was created that precisely defines which article classes are to be planned and scheduled and how. With such fundamental analyses, existing stocks can be quickly reduced and delivery readiness increased at the same time.

But all such analyses and the measures derived from them are not enough if, for example, dispatchers are not supported by suitable software. At a valve manufacturer, for example, the reporting and safety stocks had to be determined by the ERP without system support. It was particularly noticeable in the case of safety stocks that they were calculated in different ways depending on responsibility or were merely the result of empirical values. In the planning of purchased parts, order requirements were not checked on a needs basis, but once a week. Despite the ERP system, the overall process at this point was therefore highly manual, very time-consuming and therefore prone to errors despite the utmost care. In order to be able to use all the “big data” from Supply Chain 4.0 at all, many series and variant manufacturers must first structure and optimize the scheduling processes upstream of the finished goods warehouse or the last production machine before packaging. And this is no trivial undertaking.

Complex matter

The complexity of scheduling can be seen from the amount of master data required alone: Depending on the cut of the item, you have to take care of up to 130 logistical parameters. If you imagine these as a mathematical equation, it is easy to understand that you cannot calculate them in your head. However, major mistakes are made when individual parameters are combined for the sake of simplicity. For example, safety stocks for fluctuating demand, safety stocks for fluctuating production times and safety stocks for fluctuating delivery times of upstream suppliers are mapped in a common safety value. Cumulatively, this can only lead to more stock. Highly oscillating graphs with many different peaks turn into curves until you finally arrive at a “smooth” forecast, which, however, only covers up the problems and ends up costing a lot of money. Optimal scheduling therefore also requires correspondingly differentiating tools.

ERP alone is not enough

Most companies already have a suitable software tool for scheduling purposes: the existing ERP system or corresponding extensions. However, ERP systems originally have other tasks, meaning that the options for demand forecasting and replenishment are usually very limited and these functionalities are not sufficiently differentiated. For example, there are practically no automatic mechanisms for the continuous optimization of MRP parameters. In addition, virtually all known ERP systems work exclusively with statistical methods that assume a so-called “normally distributed” demand, such as mean value methods or exponential smoothing. In practice, however, normally distributed demand is practically never encountered. Rather, demand is subject to constant seasonal, cyclical or other fluctuations. As a result, calculations based on the assumption of normally distributed demand lead to systematically incorrect demand forecasts and inventory errors of up to 40%.

Trade article, Baumarktmanager, published in issue 6/16ByAndreas Capellmann and Andreas Kemmner