Coordinating a Dynamic Logistics Network in Motion
With around 1,300 stores and a 36% market share, ICA Sweden (ICA) is Sweden’s leading grocery retailer. Its cooperative model combines national scale with local flexibility, as independently owned retailers operate their own stores and tailor assortments to local demand. Each store therefore functions as a customer of ICA’s central logistics organization, making stability and predictability essential as even short-term delivery disruptions risk eroding store-level trust.
ICA’s logistics network includes six distribution centers and one e-commerce fulfillment center, supported by approximately 5,000 employees delivering goods to nearly every Swedish municipality. In addition to food retail, ICA manages ICA Special, which sources and distributes non-food products sold through Maxi ICA Stormarknad and wholesales to ICA stores across Sweden, adding further operational complexity.
As ICA embarked on a major logistics transformation, they needed to relocate large flows across several warehouses in Helsingborg, Kungälv, and Västerås while keeping store deliveries uninterrupted. Each transition required redistributing product flows across sites, often in multiple phases. Historically, ICA had no reliable way to transfer inventory freely between distribution centers without disrupting forecasts or increasing operational risk. Any major warehouse change risked triggering long-lasting downstream effects for stores, which ICA was determined to avoid.
Channel-Level Forecasting and Seamless Warehouse Transitions
A RELEX customer since 2017, ICA worked closely with the team to develop a forecasting approach that could support a logistics network in constant motion. Together, the teams introduced channel forecasting, which predicts demand by sales channel rather than tying forecasts to fixed warehouse structures.
By decoupling forecasts from physical distribution centers, ICA gained the flexibility to reassign volumes and transfer inventory as warehouse roles changed. Historical data could be recast to reflect new logistics configurations while preserving underlying seasonality and demand patterns, allowing transitions to be planned in advance rather than managed reactively.
The approach maintained forecast quality even as distribution centers were upgraded and phased in and out of operation. It also made it possible to generate reliable forecasts for products and flows with limited or no historical data, removing the need to rebuild forecasts or rely on newly created sales histories. Without this capability, each warehouse change would have required manual forecast rebuilds, increasing error rates and buffers, and raising the risk of delivery changes being felt by stores.
Results and Outcomes: Stability at Scale
Channel forecasting allowed ICA to carry out large-scale warehouse relocations as well as upgrading automation without disrupting store operations or degrading forecast quality. Forecasts remained stable and usable even when products were reassigned across sites or introduced into new flows with limited historical data. This made logistics changes plannable and predictable during a period of significant network transformation.
Most importantly, participating stores experienced no delivery disruptions or service deterioration. Because ICA operates a network of independently owned retailers, maintaining trust and continuity at the store level was a core requirement. Throughout the project, stores remained unaffected despite extensive changes to the underlying logistics network.
Given ICA’s role as a major contributor to Sweden’s national food supply, avoiding disruption was the primary measure of success. Rather than focusing on short-term efficiency gains, ICA prioritized and succeeded with stability and risk avoidance.
Smarter Allocation, Stronger Partnership, and Continuous Improvement
For ICA, logistics optimization is an ongoing effort. As new warehouses come online and automation expands, the ability to plan confidently through change remains essential. RELEX continues to support ICA by working closely with its teams and developing solutions aligned to its operating model as new challenges emerge.
From ICA’s perspective, the value of the project lay not only in the solution itself, but in how it was developed. The challenge was complex and not fully defined at the outset, requiring close collaboration and a willingness to rethink how forecasting could support large-scale logistics change.
According to Tom Söderberg, Head of Supply Chain Analytics at ICA Sweden, this approach made a difficult transition manageable.
“We were facing a situation where we needed to change large parts of our logistics network, but we didn’t have a clear way to do that without creating downstream issues. Thanks to close collaboration with RELEX, we found an approach that allowed us to maintain stable forecasts and avoid negative consequences for our stores during a very complex transition – maintaining a sense of ’business as usual.’”
“What stands out about RELEX is their attentiveness and willingness to evolve with us,” said Andreas Persson, Director of Replenishment at ICA Sweden. “Over the years, we’ve grown together, always aligned in our goals, values, and ways of working. It’s a partnership built on trust, shared ambition, and a strong cultural fit. Whenever new challenges arise, we approach them collaboratively and find solutions that create lasting value across our network.”
With RELEX, ICA has established a forecasting foundation that supports adaptable and predictable logistics operations. By solving a complex planning challenge together, ICA and RELEX have shown how large-scale network change can be executed with confidence, without disrupting service to stores.