How Advanced Analytics is Revolutionizing Retail Supply Chains

Some of the key pain points I have observed in my time at Octave anchoring supply chain use cases in the retail domain have been having the right balance between stock-outs and overstocking. The era of advanced analytics means that retail businesses have a more significant data landscape to try and eliminate these pain points. In this short article, I will walk you through the biggest transformations advanced analytics can bring to supply chains — both the learnings and challenges.

What is Supply Chain Analytics?

Advanced analytics uses a variety of data sources embedded into data science models and analytic dashboards to provide insight and decision-making to retail supply chain teams. Some of the use case applications within the retail industry are;

· Demand forecasting — Predictive analytics can leverage hundreds and thousands of granular transactions, sales and economic datapoints to uncover hidden relationships and trends to provide more accurate and granular demand forecasts to demand planning teams. This includes being able to accurately predict demand during promotional periods as well as seasonal periods.

· Inventory planning — Combining demand forecasting, supply chain teams can now merge their order management systems with automated forecasting, along with service level stocks to reduce stock outs within outlets at an SKU-level.

· Key alert generation — By using real-time analytics, demand planners as well as outlet managers can identify when certain SKUs are running out of stock at warehouses or outlets respectively. This ensures that corrective action can be taken to reduce lost sales due to out-of-stock.

· Supplier management — Advanced analytics can evaluate supplier performance using past data on fill rates and delivered quantities to create supplier reliability scores. This ensures that any risks to the supply chain are proactively identified and mitigated.

What are the key challenges and how to overcome them?

Within the Sri Lankan context, there are multiple issues that can disrupt the smooth flow of advanced analytics operations within the supply chain.

· Event-driven risks — Sri Lanka is no stranger to event-driven risks that not only affects retailers within the country, but all businesses across the island, such as political and economic unrest. While there are safety stocks in place to eliminate this, this results in high loss margins or stock holding costs for the business due to the unanticipated nature of these events.

How to overcome it?

Forecasts need to provide an additional service level stock based on the uncertainty of demand. Further, both analytic and business teams need to partner up during these times to decide if forecasts need to be overwritten as models cannot anticipate economic unrest.

· Data Silos — Supply chain forecasts are often required in multiple facets of retail businesses, such as for promotion planning by commercial and marketing teams. However, different departments within organizations work in isolated environments, and merging these processes analytically provides a challenge.

How to overcome it?

Creating a use case for the supply chain team does not involve only operating with that specific team. In the ideation of the use case, all relevant stakeholders must be gathered to identify how to digitize/roll out a use case across all related domains.

· Organizational resistance to change — Most retail giants within the country have been operating their supply chains long before advanced analytics took over. Therefore, getting buy-in from leadership teams as well as on-the-ground demand planners for a completely new system often takes time.

How to overcome it?

Constant and steady communication with business teams will significantly increase buy-in. By running pilots to evaluate impact on accuracy, operational and financial metrics, and taking business input into feature engineering, business teams will be more willing to use analytic-driven approaches.

To conclude, the above-listed use cases are a mere fraction of what advanced analytics is capable of doing for the retail industry. Newer technologies, such as generative AI, have also started taking supply chains by storm. Transforming supply chains mainly requires two enables — capabilities and an environment willing to test and deploy these solutions.