Supply chain processes and transactions have been captured and automated by IT solutions. The process automation will help the supply chain planners and practitioners to do and track the operation tasks easily. While these innovations would reduce the planners daily job hassle significantly but not necessary would help planners to get the most efficient and optimal solutions.
The planner requires a way to translate operational requirements and constraints into something that computer can understand and use to produce not just a solution but an effective solution. Let’s call this requirement, an operational model. For example, the shipment of products into loads of a truck is an example of loading model. A very simple model uses product’s weight and volume as a loading requirement and produces an efficient load profile for shipment. One might ask, why this efficiency is matter and how we could gain this efficiency.
For a company that ships billions of pounds of material each year to thousands of customers, shipping costs are a significant components to the cost of finished goods. Thus, small improvement on the shipment efficiency easily will add up to a significant cost saving and provides strong competitive advantage for the company. Now, let’s see through a small example how the shipment could be improved.
Let’s assume a transportation manager currently deals with a constraint of weight and volume. He loads one truck to max weight and load another to maximum number of pallets. Although this load profile is OK when considering one truck at a time, since we reached to one of our constraints, but in reality the first one has more room and second one could take more weight. It is apparent that the loading profile could be improved and by doing so more items could be shipped in one order.
The calculation of the best loading profile is non-trivial, assume one shipment and loading requires several trucks, the planner could spend enormous amount of time to come up with the best plan or close to the best plan, however, a computer through a model could solve this complex problem easily and provides the solution to the planner in manner of seconds.
The model development needs a special attention. In supply chain, there are a lot’s of variability which often leads to errors in model result and poor supply chain decisions and easily could disappoint the decision makers. Therefore, there are some crucial elements in developing a model that must be observed. First the data must be accurate and comprehensive. For example the weight and volume of the items for given example must be accurate and up to date. Second the process must support optimization and ability to improve continuously. It simply means the process requires systematic monitoring of data, models and algorithm performance to support and continuously improve supply chain and logistics optimization.
Nonetheless, if the optimization models are developed according to the operation requirement and integrated with supply chain operational process then it would provide an opportunity to reduce the operation cost by 10% to 40% through an efficient and better decisions.