Advanced Analytics Requirement in Inventory Management
Meet the Experts
⇨ Inventory visibility in itself is not sufficient. Leveraging that visibility for optimal inventory planning is critical.
⇨ Advanced analytics plays an important role in translating visibility into actionable insights.
⇨ SAPinsiders in response to our survey on inventory management and planning, have selected advanced analytics as one of the top three requirements to build robust inventory management and planning capabilities.
SAPinsider will publish a benchmark report on Inventory Management and Optimization in November 2022. There could not have been a better time to conduct research focused only on inventory management and optimization. Why so? Well, the first aspect, which is not surprising to any of us, is the immense focus on supply chain and inventory management that we have seen in the last couple of years. As it applies to critical industries, inventory management is even being discussed in the White House. So that provides an excellent opportunity to explore this topic, to get more insights on this topic for executives so that they can leverage that to harness the energy that has built around transforming inventory management, put it to better use for getting funding, through getting business cases approved and, using that energy positive momentum to transform their inventory management process. In this article, we will discuss why SAPinsiders have chosen advanced analytics as one of the top requirements for building robust inventory management and optimization capabilities.
The adoption of advanced analytics methodologies in supply chains has increased recently, but many areas within supply chains are still using approaches that may not be optimal in this era. We have seen the global havoc wreaked on supply chains during the pandemic because of the just-in-time approach. Another approach that leads to a struggle for many companies is the rigid safety stock calculation approach at the core of traditional inventory optimization tools and methodologies. Safety stock provides a decent grasp of how much inventory you need to hold across your supply chain. However, it may not be sufficient by itself as far as today’s dynamic business ecosystem goes. As supply chains have developed and become much more complex in the last couple of decades, it may be the right time to use other analytics levers with traditional inventory optimization approaches to building more realistic inventory management and planning processes. It is, therefore, not surprising that SAPinsiders placed advanced analytics among the top three requirements for building robust inventory management capabilities.
While inventory optimization remains one of the key levers for inventory management and planning, below are some additional approaches that can help you ensure optimal inventory levels in your network. Please note that these approaches need to align with each other to implement a truly optimized inventory planning process.
Demand forecasting: While it is a simple decision that accurate demand forecasts play a significant role in helping reduce on-hand inventory, traditional time series methods have struggled as the complexity and noise in demand data have increased exponentially. Machine Learning (ML) based advanced analytics algorithms can tackle many challenges traditional time series forecasting algorithms are running into.
For a detailed overview of using Machine Learning algorithms for demand forecasting, please refer to this article: Leveraging Machine Learning for Demand Forecasting.
Supply chain segmentation: Many organizations have traditionally leveraged a “one size fits all” for inventory planning, where the traditional inventory optimization method was force fit on local or regional supply chain demand data. But today, many supply chains operate within a single supply chain, even in the same geography. Optimal data-driven supply chain segmentation is a must before embarking on an inventory optimization journey. Advanced analytics algorithms, like k-means clustering, can help you automate your supply chain segmentation exercise, thereby reducing your planning cycle time for inventory planning.
Strategic inventory classification: For decades, organizations have traditionally leveraged ABC classification for inventory categorization. While it is still a better approach than any other tribal approach, the realities of product portfolio and demand dynamics are much more complex in today’s supply chains than when ABC analysis first came into existence. Advanced analytics can help here as well. Clustering algorithms can help create more strategically aligned clusters (categories) vs the classic ABC approach. While this may increase the number of inventory groups, the potential to reduce inventory costs is significant.
Optimal network design: It should not be news for any supply chain practitioner that the network footprint of your supply chain affects the amount of inventory you hold in your network. Organizations have been leveraging supply chain network modeling to design optimal supply chains for decades, but not all such exercises optimally incorporate inventory cost impact. If you do not incorporate inventory cost aspects into your network design, you are essentially “locking” avoidable inventory costs every time you redesign your footprint.
Manufacturing optimization: Leveraging even vanilla simulation tools can help you minimize the inventory in your manufacturing process and raw material inventory. Suppose your supply chain systems are optimally integrated. In that case, you can incorporate dynamic data points like inbound raw materials and semi-finished goods data and demand pulls into your manufacturing planning to optimize your manufacturing planning even further. And this affects the total inventory you hold in your network.
Warehouse optimization: Since warehouses perform the ugly task of holding the inventory, they are always the focus of inventory optimization initiatives. But most of the time, the focus is on finding the optimal inventory level to hold at each warehousing location based on the demand that the particular warehouse fulfills. However, inventory reduction opportunities hide in warehouse layout and flow design. Advanced analytics can help design optimal warehouse layouts, and leveraging analytical approaches to data captured by WMS systems can help you run a leaner warehouse.
Transportation optimization: Ever since the manufacturing footprint of companies started getting globalized, transportation has played a crucial role in contributing to the amount of inventory held in the network. Leveraging advanced optimization algorithms and heuristics, you can design transportation networks and routes that are “inventory friendly,” which means they try to minimize the impact of transportation on the network inventory.