Most pharmaceutical companies use Sales and Operation Planning systems in order to perform long term planning of their production. This may be helpful to give some level of visibility by high level planning but how does it help in the following areas?
- Optimizing the use of resources and ROA
- Increasing Inventory Turns
In order to optimize resources and ensure the right product mix, the system needs to be capable of modeling the environment for tanks and packaging. It needs to take into account set-up times, campaigns, product mix and synchronization with packaging lines. It also needs to have the ability to interact with the shop floor system to generate schedules daily or by shift. If anything goes wrong the system would re-evaluate the plan and based on the current constraints and current and future demand, generate a new plan in almost real-time, making sure that the utilization of resources is optimized. Using this approach, Adexa has implemented solutions that can automate the planning process and help the planners to engage in higher levels of planning activity instead of tactical assignment of resources.
using predictive Machine Learning techniques, one can predict the level of inventory needed depending on many factors, not just time of the year or region.
Inventory, in almost every industry, is a necessary evil. The only question is how much is enough. Most companies try to minimize inventory by improving their inventory turns. However, this measure does not reflect the potential impact of inventory on revenue. Too little inventory may cause missed opportunities and too much inventory would lead to excess cost and reducing profits. Inventory maybe undesirable however it is both an investment as well as insurance to protect the business. Uncertainty is given and inventory helps to deliver when demand changes or disruptions occur. Keeping the wrong inventory is bad investment and a high premium to pay for your insurance. Depending on the industry, companies have different policies regarding how much inventory to keep and where in the supply chain it should be kept. The common thread is the question of responsiveness, which is a choice between leadtime to make and inventory to keep. For example, semiconductor and pharmaceutical industries have long leadtimes to make and cannot afford to lose market share, hence they are inclined to keeping, perhaps more than, enough inventory. For industries where there are too many configurations and options keeping FG inventory is virtually impossible. They need to be in a position to react and respond very quickly. To do so, they have to keep the right level of partially made products along the supply chain. Such that they can reduce leadtime to delivery of the final product. This can also be done in pharmaceutical as well as semiconductor industries. For example, API can be ready for packaging in pharma industry as semi-finished wafers can be buffered for additional layers in semi industry.
For most industries, the companies have to walk a fine line between service level and cost of inventory. But having a high service level does not necessarily imply a higher cost! One can deploy postponement strategies so that for a given service level lowest amount of inventory is deployed across the entire supply chain. This can be done using stochastic techniques (MEIO) to predict how much inventory should be kept at each echelon of the supply chain so that based on specific products or customer(s) the accepted leadtime is adapted at the lowest cost. Also, Product life cycle and other attributes of the product can play a role in how much inventory is needed. Learn more about how stochastic learning techniques can be deployed to reduce your inventory by as much as 30% Here.