Challenges of Excess Inventory Management Have Increased

Inventory planning and management have taken a whole new form in the COVID era. Optimal inventory planning and management were never easy, but the challenges have now increased manyfold. As organizations changed their inventory strategies, many of them moving from just-in-time to just-on-case, the associated planning, and management challenges also need to be addressed. In this article, we discussed the recent inventory-related challenges that some leading retailers have run into. The specific scenario here was excess inventory. Our postulation was that with the right combination of people, processes, and technology, companies can manage these scenarios better. This article will discuss how AI can help address these scenarios from a supply chain planning perspective. We will then explore a solution from a vendor partner that has a feature that leverages such capabilities.

Artificial Intelligence (AI) To the Rescue

If we simplify the problem significantly, the challenge, or the problem statement is that you have more product(s) than what will eventually get consumed. This means that if you wait till the end of your selling time horizon to find out how much excess inventory you have, either you end up with excess inventory that you will need to sell at a heavily discounted price, or your inventory may become obsolete, with no salvage value at all. Now there are two approaches to manage this, both of which are necessary. One is preventive, where you leverage historical data, forecast, and other internal and external data points to plan your merchandise and optimal buy quantities and cycles. The second approach pertains to planning, the proactive approach, is where you keep a tab on your inventory to proactively identify inventory that can run into the “excess” zone. In this approach, you also use the demand consumed, forecast for the remaining time horizon, historical consumption pattern for the same or similar product(s), shelf-life (if applicable), and additional external and internal data points to determine if you will have excess inventory. And AI can play a big role in helping execute both these approaches. In this article, we will explore the proactive approach. Proactive approach As mentioned in the previous section, the proactive approach means that you continuously monitor your inventory, and consumption rates, and forecast and adjusted demand. As figure 1 suggests, it is humanly not possible for a planner to incorporate all these data points simultaneously to formulate an optimal excess inventory management strategy. Irrespective of the level of manual effort put into the process, the results of the manual analysis will most likely be non-optimal. And hence, this makes a perfect use case for leveraging AI. Figure 1. Multiple data points are required to formulate an excess inventory strategy This is where I would like to introduce a solution from a vendor partner that I recently came across. At a high level, you can categorize INTURN as an inventory planning and management solution, but I love how they phrase their offering “INTURN addresses the largest ongoing problem for brands: unplanned inventory and turns it into a business opportunity.” Excess inventory obviously is at the top of the list among the list of inventory problems. And that is why, a unique capability of their INTURN 360 solution is an AI model that can proactively predict inventory at risk of becoming slow-moving or excess, as well as optimize existing overstock to maximize recovery and streamline workflows. As INTURN's co-founder, Charlie Ifrah highlights: “Companies face an enormous financial burden when dealing with unexpected levels of excess inventory—higher warehousing costs, minimal margin recovery, and reallocating resources to manage that inventory when they could be used for other profitable areas of the business. By leveraging AI to predict inventory at risk of becoming slow-moving or excess, teams can understand the most effective way to offload that inventory earlier in the product lifecycle."

What Does This Mean for SAPinsiders?

As highlighted in a previous article, technology solutions need to be paired with other two components, people, and processes. As solutions like the one described here present recommendations, you need to define a process around how you will act on these recommendations. Here are some recommendations that can help you define the process of how to react to recommendations: Train your planners. As Figure 1 suggests, it is not realistically possible for your planners to account for all the data points to make optimal excess inventory analysis. And this is where technology, in the form of tools like the one described above, comes into play. But there are two key aspects that will translate into the tool actually becoming an enabler. Thousands of planners in enterprises across industries secretly use their own excel templates to override inventory management decisions from highly-priced sophisticated tools. And the key reason is the lack of trust in data and the tool methodology. This is where training becomes really critical. Understanding the tool methodology instills confidence in recommendations and will also help the planners fine-tune the solution if they believe it needs parameters tuning to generate more robust recommendations. Become best friends with your data points. The Garbage-In-Garbage-Out (GIGO) principle applies here as well. But even if your data points that feed the toll are robust, your planners need to know the data inside-out. Often, having the "feel" of the data points that go into the tool helps you decipher the recommendations. As an example, though you may not have the exact numbers and recommendations like the tool, an SKU being highlighted as being probable of being running into the risk of excess should not come as a surprise to you (most of the time). That "feel" of data points is essentially your data-driven "feel" of business, your business intuition. Form a task force. Excess inventory is a regular nuisance in many industries. An element of putting a process together is a group of stakeholders who will act upon recommendations or data from technology solutions. These stakeholders will be from multiple functions, from teams like merchandising, marketing, supply chain planning, store operations etc. Though these stakeholders need not be in-depth with data points and the tool as the inventory planner, they still need to be fairly conversant with how and why the recommendations are being generated.