Slow moving and intermittent demand items present unique forecasting difficulties for companies. Mark Watson argues it is not an impossible job and, if perfected, can deliver significant benefits to an organisation.
Finding a solution to the challenge of deciding what stock to hold for slow moving and intermittent demand items is often regarded as the holy grail of supply chain management, as companies try to find the balance between service delivery and cash flow.
The slow moving or intermittent demand item often has no discernible usage pattern, yet when combined they often form a large part of the product range that companies utilise on a day-to-day basis. Typically decision makers around inventory holding may take the safe option, which sees them stock one of everything ‘just in case’.
The downsides to this approach are clear. Often the slow moving item is large and expensive, such as a generator or a gearbox; it has a long lead time, takes up considerable space in a warehousing facility and ties up large amounts of working capital.
Faced with this problem the logical approach is to turn to the world of information technology to help, seeking support from an off-the-shelf software package or by creating an in-house database in a bid to try and seek out meaningful patterns in the demand that is being experienced. There are a number of challenges in doing this.
Finding patterns in the unpredictable
Firstly, the nature of this demand, by definition, offers little data from which accurate demand computer models can be formulated.
Secondly, every business is different and there are multiple factors that can affect demand. For the automotive aftermarket, new model launches or model facelifts are a major factor, while for the Ministry of Defence, training, redeployment and conflict can affect demand for parts and equipment.
However, while these multiple factors are often difficult to predict, companies can rest assured that order can be brought to what often appears to be chaotic and random.
In order to accurately forecast, there is a process that companies can go through. The first step is to differentiate between slow moving items and intermittent items. A slow moving item has average demand below a certain value, whereas intermittent demand has the average interval between demand occurrences greater than a certain value.
For example, slow moving stock for an automotive parts supplier could be defined as items for which average annual demand was less than 20 pieces, with intermittent demand stock defined as being items for which the average interval between demands is greater than two months.
Secondly, it is important to acknowledge that standard computer algorithms don’t provide the necessary differentiation between different categories of items. Slow moving stock demands very different procedures due to the nature of the demand for these items, violating the assumptions for which conventional forecasting algorithms are designed. [sam_ad id=6 codes='true']
Finally, it is vitally important to embrace the ever-changing nature of a part through its life cycle. For example, within the automotive aftermarket, there is often clear demand increase once the parc includes large numbers of vehicles with more than 50,000 miles on the clock. In this situation it is important to successfully manage the transition of a part from slow moving to fast moving in order to maintain service levels.
Practically, the output is two sets of calculations, a forecasting method and an inventory control method produced either daily or weekly depending on the nature of the client. The forecasting method uses demand history to calculate average demand over a period. The inventory control method can then take these estimates to produce schedules for replenishment orders, which can be used to manage the flow of material through the supply chain.
It is important to note that no inventory management system will ever deliver 100% availability. Certain items will just never be stocked and will always need to be ordered from the supplier in the unlikely event that they are required by the end user. The key to success is about being realistic with your inventory control parameters, the results they can deliver and that some items will always be impossible to forecast.
Every organisation is different, however, it is possible to deliver in the region of 10-15% inventory reduction without affecting service levels. Although it can be as much as six months before these gains are realised, our experience suggests there is clear evidence that significant cost reductions can be made.
In addition to the financial benefits realised from not having capital tied up in excess or slow moving stock, the longer term benefits include releasing warehouse space from slow moving items or preventing excess wastage when products become obsolete.
The truth of the matter is that technically minded inventory policy is possible and can make significant contributions to the bottom line of an organisation by reducing stocks, improving working capital and taking cost out of the supply chain.
Mark Watson is supply chain and operations director at TVS Supply Chain Solutions