inventory and supply chain management

Vinod should consider forecasting all SKUs. Less than 3% of the SKUs were normally distributed. Without forecasting, there is no way to correct for the errors leading to Dockomo's current problem. If it helps, Vinod might be able to leave the SKUs with normal distributions alone, like the Cr106, Cr137, and Cr108. Otherwise, forecasting will help create a more effective supply chain distribution to meet the targeted service delivery goals.

Exhibit 6 also shows that the percentage of stockouts does not vary overly much between the categories (V, M, S, and VS). This suggests that forecasting is needed for all categories and not just a few. In 2010, the company recorded a forecasting error of 70% for the top 5% of items, the ones contributing to the majority of their revenue and profit (p. 8). Therefore, Vinod might want to consider focusing on these top 5% at first, because of the clear connection to revenue and profit.

With a target of customer service improvements, though, and the goal of 95% fill rates, just basing the forecasting on which items are costing the company most will not necessarily be a good idea. Vinod needs to address the inventory and supply chain distribution errors for all products. Basing orders on experience is fine for small companies with less complex supply chains, but not in a situation like this, in which demand is erratic, and supply is also variable. 2.

Creating safety stock is a natural reflex to preventing future problemsin enterprise resource planning, and Dockomo should consider reducing its safety stock in accordance with the results of statistical analyses of both supply-side issues and demand forecasting. Currently, the company relies on anecdotal evidence and the prior experience of employees, which is barely better than intuition in helping to create a realistic inventory that meets actual customer needs, while also taking into account space limitations.

The lack of updated inventory or buying policies, coupled with a mushrooming master parts list, is a recipe for disaster. "The tendency is often to hold too much inventory, and thus to avoid stockouts and the resulting fallout that lands on inventory managers," (Silver & Pyke, 2016, p. 806). A safety stock that is 1.2 to 2 times customer demand is unsustainable (p. 8). The safety stock problem is one of methods. Normal distribution only applies to three items (Cr106, Cr137, and Cr108).

The rest of the items, comprising the bulk of Dockomo inventory, needs to reflect a methodology allowing for skewed distributions. In Dockomo's case, the distributions were extremely right tailed, all but the three normal SKUs and the two left-skewed SKUs (Cr108 and Cr1370. Using inventory management software, Vinod can input the data for each SKU, and come up with an appropriate scheme for managing safety stock.

Examples of variables that can be input into the algorithm for optimal prevention of stockouts would include order data, shipment data, and existing inventory. 3. Vendor EJK from Korea has lead times of three months (machine parts) and four months (hydraulic parts). EJK is a critical partner, being the sole vendor for critical machines and parts for the most popular crane models: the EX300, EX270, EX150, and EX70 (p. 7).

Accommodating for the vendor lead-in time should not be too difficult if adequate software and exponential smoothing are used, and yet in this case, the research showed that "very few SKUs followed time-series characteristics, such as trends or seasonality," (p. 9). Although it can be difficult to forecast demand three months ahead, Vinod will need to at least base the forecasting on some past data to see if there are seasonal trends in the business.

The 20% of SKUs that can be predicted based on trends and seasonality should be predicted accordingly. The remaining 80% is the real problem. Relying on experience is fine for some things, but not in this case, where the items are costly and demand is lumpy.

It is to be presumed that all other competitors will be having a similar problem; if EJK is the only provider of the critical parts, then that means the onus is shifted on the client for communicating needs as soon as possible to avoid time delays in their projects. For example, Dockomo can communicate that this item is produce on an as-needed basis, and pass on this information to clients.

If Vinod can ensure that communications channels with clients remains absolutely open and clear, with frequent follow-ups and check-ins, then it becomes possible to at least avoid the disappointments that have ensued and ensure customer satisfaction as well as ongoing brand loyalty. 4. Estimating the reorder point would also require modeling, such as stochastic modeling (Glynn, n.d.).

The modeling algorithm would have to take into account numerous variables, including the SKU category (fast movers, medium movers, slow movers, and very slow movers), which would be one of the most important means of determining the reorder point. Vendor lead-in time would also be important to factor into the.