John E. Boylan

Intermittent Demand Forecasting


Скачать книгу

alternative calculation, which attempts to address this issue, is to multiply the forecasted demand per week by a period that is longer than the protection interval, to allow for demand uncertainty. For example, we could multiply the forecasted demand, of two per week, by four weeks, instead of three, to give an OUT level of eight units. The problem now is that the setting of four weeks is arbitrary. Why not use five or six weeks instead of four? There is no guarantee of hitting service level targets using this type of calculation. For highly unpredictable demand, we may set the OUT level too low and, for more predictable demand, we may set it too high. Therefore, although this rule of thumb has the merit of simplicity, it risks service level targets being missed or targets being achieved with excessive stocks.

      3.2.2 Judgemental Adjustment of Orders

      Although the experimental evidence on judgemental ordering is not encouraging, there are situations when it can be beneficial. A prime example is to take advantage of discounts from suppliers that are available only for a limited period of time. In making a judgemental adjustment to the order, the demand forecast should be left untouched if no change in demand from end‐customers is anticipated. In other circumstances, it may be appropriate to adjust the forecasts themselves. We shall return to forecast adjustments in Chapter 10, where there is a detailed discussion of the subject.

      To summarise, judgemental adjustment of orders may have positive effects in gaining price discounts but can also contribute to the bullwhip effect, leading to additional inventory holdings. The effect of order adjustment may not be as great as the impact of forecast adjustment, as shown by a system dynamics investigation (Syntetos et al. 2011b) and a further empirical study (Syntetos et al. 2016b). Nevertheless, any price benefits should be weighed against increased inventory costs. If order adjustments are frequent, it would be good practice to keep records of the original recommended orders, as well as the adjusted orders. In this way, an organisation can monitor the effect of judgemental adjustments and evaluate whether they are beneficial or, more specifically, the circumstances under which they are beneficial.

      3.2.3 Summary

      Order‐up‐to (OUT) levels are sometimes set by using a rule of thumb of multiplying the forecasted demand (per period) by a set number of periods. This ignores the degree of demand uncertainty, leading to a misdirection of inventory investment. Consequently, rules of thumb like this are not recommended.

      Judgemental adjustment of orders may be worthwhile in some situations. These adjustments should be continually monitored, to evaluate their effects on price benefits and inventory costs.

      Organisations should assess their inventory performance with respect to financial and service targets. While service targets are needed at SKU level, assessment of service targets at higher levels of aggregation elevates an operational task to a strategic issue. A strategic view of service and costs is essential because there is often a trade‐off between these two aspects of performance and looking at just one aspect may lead to poor decisions.

      3.3.1 Aggregate Financial Targets

      An important issue to be addressed is the ‘write off’ of obsolescent stock if it has no residual value or its ‘write down’ if it still has some value, for example as scrap. Once stock becomes so slow moving that it is unlikely to be ever sold, then these adjustments to the balance sheet are necessary. The record of inventory written off, or written down, is useful in two ways. Firstly, it can be used to assess the additional percentage inventory holding cost to allow for the potential obsolescence of slow moving SKUs, as discussed in Chapter 2. Secondly, the write‐off and write‐down values can be monitored to assess the effectiveness of measures to prevent the build‐up of obsolete inventory.

      Inventory turnover is another useful financial measure. It is calculated as the ratio of the cost of goods sold (COGS) to the average inventory valuation. For example, if the annual COGS is £8 million and the average inventory valuation is £2 million, then the inventory turns four times a year. This measure is useful because it takes into account the growth or decline of sales. If COGS rises by 20% but inventory value increases by only 10%, then the inventory turnover will improve. It is possible to benchmark a company's performance on inventory turnover against competitors, but care is needed in drawing conclusions from the comparison. Inventory turnover is strongly influenced by the breadth of products in the stock range. If it is necessary to include slower‐moving stocks, because of contractual obligations, then the inventory turnover will necessarily be lower. If required, inventory turnover can be assessed for categories of SKUs (e.g. by volume of sales) or at the level of an individual SKU. It is most commonly measured at the aggregate level and provides a useful headline figure. It should be monitored over time, so that the effect of any changes in policies or practices can be assessed.

      A full financial assessment of the impact of stockouts can be difficult to achieve. If stockouts lead to backorders, then it may be possible to estimate the costs of expediting orders to minimise the delay in satisfying the order. It is more challenging to assess the costs of lost goodwill from clients. For this reason, aggregate financial targets are usually complemented by aggregate service level targets, to which we now turn.

      3.3.2 Service Level Measures

      Some organisations have formal service level agreements (SLAs) with their major suppliers. These usually specify target service levels and may include financial penalties for missing these targets. This highlights the importance of using appropriate forecasting and inventory management methods. Otherwise, as Willemain (2018) emphasised, suppliers will incur financial penalties much more frequently than they were expecting. An SLA will specify either lump‐sum penalties or penalties proportional to underperformance. This choice needs careful consideration as it can influence ‘gaming’ behaviour by the supplier (Liang and Atkins 2013).

      It is essential that the service level measure is clearly