Every business person who deals with inventory management is involved in a balancing act. Keep too little inventory on hand and you risk losing sales and, on occasion, customers. Keep too much inventory on hand and the costs add up resulting in lower profits. It’s little wonder inventory management systems have gained popularity. Sean Russell explains, “These services provide a supply chain that keeps tabs on products through every stage of development, and allows management to get real-time estimates of their available stock throughout the day.” He adds, “Even with all this power, there are several disadvantages to inventory control worth considering.” Although the benefits of inventory management systems far outweigh the disadvantages, Russell cautions such systems could increase bureaucracy, de-personalize customer relationships, and hide production problems. Omnichannel operations exacerbate inventory management complexity. According to research conducted by Adelante SCM and LEGACY Supply Chain Services, and published in their “Omni-Channel Logistics Leaders: 5 Key Insights to Improve Inventory Performance for 2019” report, “In order to improve their omni-channel fulfillment performance, companies need to overcome several challenges that are currently hindering their ability to effectively optimize and manage their supply chain inventory. And it begins with improving their inventory accuracy.”
Inventory and forecasting
Getting the inventory balance right relies heavily on accurately accounting for inventory on-hand and correctly forecasting product demand.
Inventory. According to the Adelante SCM and LEGACY Supply Chain Services report, “The reality is that inventory data is spread out across multiple systems, multiple trading partners, and multiple locations. Aggregating, normalizing, and cleansing this data from across the supply chain (in real time or near real time) is not a trivial task. It takes a lot more time, money, and resources to accomplish than simply investing in ‘supply chain visibility’ software. Complicating matters is the fact that many companies still rely on manual processes to capture inventory data.”
Forecasting. There are no perfect forecasts. Once that truth is accepted inventory managers can decide what forecasting method best suits their needs. Traditionally, inventory forecasts have been deterministic. They have been based on historical data and often result in a single, best estimate, or may include inventory sensitivities which provide low, base and high cases. These forecasts are not dynamic, which means sticking to them could result in poor inventory outcomes. A more dynamic forecasting method is probabilistic. Sarah Lafferty, a strategic communications expert with Round Earth Consulting, explains, “To understand the relationship between probabilistic forecasting and inventory, view your supply chain as a dynamic system, subject to uncertainty and unpredictable change. Spreadsheets and legacy suites like SAP APO produce top-down aggregated forecasts using a deterministic approach. While easier to comprehend, in this environment they produce chronically poor forecasting outcomes. However probabilistic forecasting doesn’t just create an average forecast, it identifies a range of outcomes and the probability of each of those outcomes occurring. Inventory optimization (IO) systems can then use this information to better identify the optimal inventory targets.” She adds, “A deterministic approach is particularly inappropriate for planning ‘long tail’ items such as specialty goods or spare parts. That’s because demand for long tail items is intermittent and doesn’t conform to predictions of ‘average’ demand or normal distributions. So demand details make all the difference.”
The most important objective of forecasting is trying to ensure production, inventory, and demand remain in balance. Michael Wilson, AFFLINK’s Vice President of Marketing and Communications, observes, “As companies expand, they often find themselves losing control of their inventory management. The more items there are to track, the less likely an in-depth analysis can be done on demand. While many inventory management systems can predict how much an item will be needed, they often aren’t completing an in-depth analysis.”
Artificial intelligence and inventory management
Fortunately, artificial intelligence (AI) can be leveraged to improve demand forecasting. Wilson explains, “‘Smart’ forecasting technology uses artificial intelligence and machine learning to help companies plan. Rather than having to manually adjust your inventory based on customer needs, you can use past samples of inventory data to determine patterns that indicate product demand. Even patterns such as seasonal purchasing can be accounted for, helping modify your projected demand based on past years and current market trends.” In fact, near-real time data can be used to ensure forecasts are dynamically updated. Wilson concludes, “Advanced, strategic forecasting methods can be used by organizations to reduce their product loss, anticipate their customer’s needs, and increase revenue.”
Amjad Hussain, founder and CEO of Algo.ai, believes AI-based inventory management systems are particularly important for retailers. He explains, “Retail is facing battles on all fronts. … AI-assisted technology running on big data can help optimize inventory at all levels in the demand chain. It even can predict future buying behavior, and detect and act on supply chain anomalies in a timely fashion. With the implementation of ‘smart’ warehouses, retailers are beginning to reach new levels of efficiency. In this current climate, it’s time retailers start looking at these AI-assisted, analytically rich business processes as a priority and strategic investment, rather than an optional helping hand.” Few, if any, companies can ignore the advantages offered by cognitive inventory management systems. Henry Canitz, Product Marketing & Business Development Director at Logility, writes, “Today’s powerful Artificial Intelligence infused systems enable continual and automatic updates of inventory parameters by taking advantage of real-time supply chain and market data. Finally, as opposed to early inventory optimization solutions, today’s leading solutions take the impact of variability in demand and supply into account when making inventory level recommendations.”
Wilson notes, “Poor inventory management can cost an organization both money and its reputation. When items can’t be sold at full price, the company loses revenue. When customers aren’t able to get what they want, the company’s reputation may suffer. Over time, poor inventory management cuts significantly into a company’s profits and makes it less likely that customers will come to the company when they have specific needs.” Cognitive systems, like the Enterra Supply Chain Intelligence System™ and the Enterra Intelligent Inventory Management System™, can help companies manage their inventory better. Such systems are particularly useful when things aren’t going as expected. Hussain explains, “Automated inventory management can detect anomalies in supply chain, allowing retailers to be proactive and deflect potential issues, such as by transferring more stock to the right location ahead of time.” Such actions can help both suppliers and retailers avoid chargebacks and stockouts. There are no silver bullet solutions for inventory management, but AI-based systems are getting closer.
 Sean Russell, “The Disadvantages of Inventory Control,” Bizfluent, 26 September 2017.
 Adrian Gonzalez, “The Inventory Accuracy ‘Confidence Gap’,” Talking Logistics, 6 May 2019.
 Sarah Lafferty, “Why Probabilistic Forecasting Is Better for Inventory Optimization,” ToolsGroup Blog, 13 November 2018.
 Michael Wilson, “How Demand Forecasting Helps You Manage Inventory,” AFFLINK, 3 May 2019.
 Amjad Hussain, “AI-Powered Inventory Management: A Make-or-Break Tool for Retailers,” E-commerce, 22 March 2019.
 Henry Canitz, “This is Not Your Father’s Inventory Optimization,” Supply Chain Digest, 15 November 2018.