Uncertainty in demand makes it necessary to maintain a certain level of product availability to avoid stock-outs and provide a required service level to the customers. A high level of product availability provides a high level of responsiveness, but increases costs because much inventory is held, but rarely used. On the contrary, a low level of product availability lowers the inventory holding cost but results in a higher fraction of customers who are not served on time.
So, the main objective of inventory management is to find a balance — optimal product stock volume.
Safety inventory is carried to satisfy demand subject to unpredictable demand fluctuations and to reduce product shortages. This type of inventory cushion is also called safety stock or buffer inventory. Safety stock can help the supply chain manager improve product availability in the presence of uncertainty.
Actual and ideal inventory behavior
In theory, we can use formulas to calculate the optimal number of units per order (EOQ – economic order quantity) and re-order point (ROP). The green line on the chart above reflects the inventory dynamics in the case of using optimal EOQ and ROP and can be named as an ideal inventory behavior. The ideal inventory behavior means, in this case, that all assumptions are met, i.e., demand and lead-time are constant. In real life, this is not the case. Both demand and lead-time fluctuate resulting in actual inventory behavior which is a black line on the chart above.
On the chart below, ROP is increased by safety stock. It considers an example where safety stock allows us to cope with demand fluctuations in some cases. However, in other cases there exists a backlog — the orders unfulfilled from stock on-hand.
ROP with safety stock and backlogs
The next chart shows an example where ROP is increased by an excessive safety stock (ESS). The ESS is so high that demand fluctuations would never result in a backlog which means a 100% product availability on stock resulting in a 100% service level. However, the inventory level on the last chart is much higher as compared to the previous ones, which results in higher inventory costs.
ROP with excessive safety stock and no backlogs
The question is how to calculate safety stock in supply chain to find the right balance between the inventory investment and customer satisfaction?
Right-sizing safety stock is one of the most challenging tasks in inventory management. To answer the question of how much and when we should replenish, first we must calculate the optimal number of units per order. That could be done with a purely mathematical approach, that uses linear optimization: we can consider the supply chain as a mathematical model and calculate stock based on input parameters (e.g. service level) using a set of linear equations and an LP-solver.
Though, we must take into consideration dynamics of the network operations, as well as operational and disruptive risks, demand variability across multiple echelons, and the unpredictability of omnichannel. That means that the analytical approach with simple safety stock formulas just will not work for real-life complex, lean, and agile networks.
On the contrary, dynamic simulation can represent the supply chain operations as they are, thus providing the technical ability to define the right safety stock under the defined conditions.
Simulation is the best technology available today for detailed inventory policy planning, for four main reasons:
To figure out the advantages simulation provides, and how it can help define the right inventory strategies, watch the Simulation-Based Inventory Optimization webinar.
watch the webinarSafety Stock Estimation is a simulation-based experiment in anyLogistix. It allows you to determine the optimal amount of safety stock in warehouses, considering the desired service level, and prevent risks associated with fluctuating demand. You will see how the tradeoff between the carrying cost and SLA penalties impacts the bottom line and will be able to make an informed decision.
To start the safety stock optimization experiment in the ALX software, you will need some inputs:
With all this information we can run a Safety Stock Estimation experiment. To perform it in anyLogistix you have to specify how high you want your service level to be. For example, you can define that a desired service level for your supply chain is 95%.
The result of the experiment offers optimal product stock volume to provide a predefined service level. What is important, in anyLogistix you can get a detailed and clear visual representation of any metric. For example, you can get your safety stock value for each product, factory or DC separately and visualize these results in the most convenient way for you: tables, histograms, etc. Thereby, you can evaluate the results and compare service level and costs with different inventory policy parameters. Finally, you will get properly configured inventory policies with the optimized safety stock that will maintain the desired service level (for all distribution centers and sites).
For a deeper dive into safety stock optimization with the anyLogistix software, watch the following how-to video.
Safety Stock Optimization with the anyLogistix Software
Want to try how the safety stock estimation experiment will work for your supply chain? Download anyLogistix trial version.
