Stockouts drain sales, erode brand trust, and push shoppers to rivals faster than legacy planning tools can react when demand surges, seasons flip, or logistics pinch points ripple through the network. Lowe’s moved to confront that volatility by broadening its partnership with Relex Solutions, extending beyond 2024’s allocation deployment into forecasting and replenishment on a single, AI‑driven platform. The aim is straightforward but not simple: connect decisions across stores, distribution centers, and suppliers so the same signal guides where inventory should go, how much to move, and when to act. Leadership framed the pivot as a shift from reactive corrections to continuous, precise decisions built on shared data. Automation is expected to strip out manual touches, free planners to focus on strategy, and ultimately raise in‑stock levels for both DIY and Pro customers across a complex, multichannel footprint.
From Siloed to Unified Planning
Lowe’s is consolidating in‑house tools with Relex capabilities to create a single source of truth spanning forecast‑to‑replenish. This replaces fragmented workflows where allocation rules, store ordering, and upstream vendor plans often diverged, creating handoff delays and contradictory signals. By anchoring each decision to one model and one data foundation, the company expects fewer mismatches between demand forecasts and replenishment triggers, tighter alignment on safety stock, and cleaner execution in promotions and resets. Camille Fratanduono, SVP of inventory replenishment and planning, characterized the change as enabling continuous decisions rather than batch cycles, with better visibility that moves the team from firefighting toward proactive network stewardship.
That reframing matters because home improvement demand is notoriously lumpy: hurricane seasons shift category priorities, regional weather reshapes cadence, and long product tails complicate store‑level accuracy. Under the prior approach, planners often mediated across competing systems and stale thresholds, which slowed response time when inventory needed to move. The unified platform reduces those frictions by standardizing the logic used for forecasting, allocation, and ordering. It also clarifies accountability: when stockouts occur, root causes are traced to the same data stack rather than debated across tools. The expected result is consistency at scale—fewer exceptions created upstream, and faster resolution of the exceptions that remain.
How the Platform Works and Rolls Out
Relex integrates multiple demand signals—point‑of‑sale trends, seasonality profiles, regional variation, channel mix—alongside network realities like lead times, vendor capacities, and DC constraints. That blend supports a single forecasting engine tuned to long assortments and climate‑sensitive categories. On top of the forecast, replenishment automation generates orders within policy guardrails, shortens planning cycles, and elevates only true exceptions to human review. Allocation uses the same logic to steer inventory where it will protect service and margin, while diagnostics surface whether a miss stemmed from timing gaps, safety‑stock settings, or network bottlenecks. The promise is prescriptive: not just alerts, but prioritized actions with suggested quantities and timing.
The rollout follows a phased cadence toward full implementation by early 2027, with integration steps sequenced to minimize disruption and enable learning loops. That approach builds on earlier AI use cases, including SKU rationalization referenced on a recent earnings call, and slots into a broader supply‑chain modernization agenda. The move aligns with a visible industry arc: Guitar Center deployed Relex for distribution center forecasting and replenishment, United Natural Foods, Inc. extended the platform across several DCs, and Wawa adopted machine‑learning‑based forecasting and replenishment for fresh foods. These concrete adoptions show convergence on integrated suites that deliver real‑time visibility, consistent logic from forecast to order, and exception‑focused operations without expanding headcount.
What This Meant for Teams and Customers
For inventory teams, the platform shifted attention from repetitive ordering to upstream planning with merchants and suppliers, where earlier signals can reshape vendor production, align lead times, and stabilize promotions. Planners engaged more with policies—service targets, lot sizes, min‑max rules—because the system executed those rules reliably and surfaced outliers that warranted judgment. That reallocation of time improved productivity by cutting noise while preserving oversight, and it encouraged tighter collaboration with DC operations to balance flow constraints against store demand. For customers, more consistent on‑shelf availability reduced unplanned trips and substitutions, while Pro shoppers benefited from greater certainty on high‑use items and project‑critical categories.
The practical takeaway for operators watching this shift had been clear: start by standardizing data and policies, then layer automation where forecast signal and service goals are strongest. Pilot allocations or replenishment policies in constrained categories to validate guardrails before scaling. Use root‑cause analytics not as postmortems but as design inputs for vendor terms, safety‑stock formulas, and DC slotting. Finally, temper ambition with sequencing—connect the planning spine first, then expand into adjacent workflows like promotional planning or omnichannel fulfillment. With that playbook, the path to fewer stockouts and steadier service levels looked achievable without burdening teams with more manual work.
