by Kumar Singh, Research Director, Automation & Analytics, SAPinsider
Inventory planning- The constant balancing act
Inventory planning is one of the most critical and most challenging aspects in supply chain planning. This means that implementing an effective inventory planning and management strategy is crucial for an organization to stay competitive in today’s dynamic global supply chain landscape. There is always room to bring more science into the planning process for this critical balancing act between service levels and the cost of inventory. With advances in data science and the proliferation of advanced planning algorithms in the world of business, there may be an opportunity to leverage a more holistic inventory planning process, one that leverages multiple advanced analytics approaches in tandem.
Advanced analytics has the capability to re-invent inventory planning and management
In my opinion, a key strategy to successfully exploiting your investment in digitalizing your business processes, is to leverage the data that is generated to gain insights and build analytics capabilities. That is the true competitive differentiator obtained from investing in digital technologies. Data is the new oil but just like oil, it needs to be transformed into a state that enhances its value. And that state is building Business Intelligence (BI) and advanced analytics capabilities from your data.
The exponential progress in data science and computing has finally started making inroads in the supply chain world. While conventional operations research has been used in supply chain and manufacturing for decades, advanced analytics methodologies are also seeing an uptick in their adoption in the supply chain world. With Industry 4.0 capabilities on their mind, companies are making a push to build advanced capabilities within their supply chain as a foundation and advanced analytics is one of those capabilities.
The need to evolve beyond vanilla inventory optimization
Adoption of advanced analytics methodologies in supply chains has definitely increased recently but many areas within supply chains are still using approaches that may not be optimal in this era. We have seen the havoc wrecked on supply chains across the globe during the pandemic because of the just in time approach. Another approach that leads to struggle for many companies is the rigid safety stock calculation approach that is at the core of traditional inventory optimization tools and methodologies.
Though safety stock does provide a decent grasp of how much invetory you need to hold across your supply chain, it may not be sufficient by itself, in my opinion, as far as today’s dynamic business ecosystem goes. As supply chains have evolved and become much more complex in last couple of decades, it may be the right time to use other analytics levers in tandem with traditional inventory optimization approaches to build more realistic inventory management and planning processes.
Analytics levers for inventory management and planning
While inventory optimization certainly remains one of the key levers to use for inventory management and planning, below are some additional approaches that can help you ensure that you have optimal inventory levels in your network. Please note that these approaches need to align with each other in order to implement a truly optimized inventory planning process.
Demand forecasting : While it is a no brainer that accurate demand forecasts play a significant role in helping reduce on hand inventory, traditional time series methods have started to struggle as the complexity and noise in demand data has increased exponentially. Machine Learning (ML) based advanced analytics algorithms can tackle many challenges that traditional time series forecasting algorithms are running into these days.
For a detailed overview on using Machine Learning algorithms for demand forecasting, please refer to this article:
Supply chain segmentation : Many organizations have traditionally leveraged a “one size fits all” for inventory planning, where traditional inventory optimization methodology was force fit on local or regional supply chain demand data. But the reality today is that there are many supply chains operating within a single supply chain, even in the same geography. Optimal data driven supply chain segmentation is a must, prior to embarking on 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 : Organizations traditionally have leveraged ABC classification for inventory categorization for decades now. 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 vs what they were 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 cost is significant.
Optimal network design: It should not be news for any supply chain practitioner that the network footprint of your supply chain impacts 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 incorporate inventory cost impact in an optimal way. The fact is that if you do not incorporate inventory cost aspects in 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 as well as raw material inventory. If your supply chain systems are optimally integrated, you can incorporate dynamic data points like inbound raw materials and semi-finished goods data as well as demand pulls into your manufacturing planning to optimize your manufacturing planning even further. And this impacts 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 the focus most of the time is on finding the optimal level of inventory to hold at each warehousing location based on the demand that the particular warehouse fulfills. However, there are inventory reduction opportunities hidden in warehouse layout and flow design. Advanced analytics can help design optimal warehouse layouts and leveraging analyticalapproaches on data captured by WMS systems can help you run a leaner warehouse.
Transportation optimization : Ever since manufacturing footprint of companies started getting globalized, transportation has played a key role in contributing to the amount of inventory that is 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.
What does this mean for SAPinsiders ?
Establishing a holistic inventory planning process is a challenging task and advanced analytics can certainly provide significant advantages. Here are few aspects that SAPinsiders should be cognizant about:
- SAP has a robust Inventory optimization and planning capability in IBP tool. Being a SAP product, it integrates seamlessly with your SAP ecosystem.
- Unknown to many, SAP provides consulting services as well, to help those running on SAP develop best in class inventory planning and management capabilities.
- If you are looking at third party tools, the good news is that there are many options available in the market. Many of these tools are cutting edge however which one is a good fit depends on your unique business requirements.
- There is also an option to develop an inventory planning platform from scratch, that taps into your SAP and other systems, using open source programs. The critical aspect of building this platform is to ensure that you build a robust data hub or data lake. SAP data hub is also an option you can explore.
Kumar Singh, Research Director, Automation & Analytics, SAPinsider, can be reached at email@example.com