Depot Rationalization Without Compromising Service Levels: A Network Design Approach

Most depot networks were never designed. They grew one market launch, one distributor request and one regional push at a time. A decade later, the map shows 25, 30 or 40 depots, and leadership asks a simple question: can we run fewer of these and still deliver in 48 hours?

We hear that question often in conversations with manufacturers across agrochemicals, lubricants, FMCG and energy distribution. The ambition is clear: lower cost-to-serve, leaner inventory and simpler operations. The non-negotiable is equally clear: customers keep receiving their orders on time.

The answer is yes, when depot rationalization is treated as a network design decision, modeled with real order, freight and capacity data, rather than a real-estate exercise decided on a spreadsheet. This guide walks through how to do it.

Depot rationalization is the process of reshaping a distribution network, by consolidating, relocating or re-roling depots, so that the business serves the same customers at a lower total cost-to-serve while meeting agreed service levels. It goes beyond closing sites. It decides which depots stay, which merge, which become cross-docks, and how inventory and transport flow through the new footprint.

Depot rationalization vs. warehouse consolidation

Warehouse consolidation usually focuses on storage space and rent. Depot rationalization takes a network view: it balances facility cost, primary freight, secondary freight, inventory holding and delivery lead time together.

Where network design fits

Distribution network design is the analytical method behind rationalization. It uses demand locations, lanes, freight rates and depot capacities to evaluate many footprint options, then recommends the one that meets service targets at the best total cost.

Across buyer conversations, the same five patterns appear whenever depot count comes up. Recognising them is the first step to a leaner network.

1. Networks built by accretion, designed by experience

One agrochemical manufacturer we spoke with supplies domestic markets from four or five plants through roughly 24–25 depots. A lubricants brand runs close to 30 depots for about 850 SKUs. Each depot was added for a good reason at the time. Few were ever re-evaluated together as one network.

2. Network planning still lives in spreadsheets

Many teams run supply and network planning in Excel, supported by SAP exports and BI reports. Spreadsheets work for today's plan. They struggle to compare ten footprint scenarios across thousands of order lines, freight lanes and capacity limits.

3. Inventory spread thin across too many nodes

When stock is split across many depots, the same SKU can be short in one location and slow-moving in the next. One oil marketing company supplying 40+ C&F depots described depot congestion, vehicle detention and excess slow-moving stock sitting alongside shortages.

4. Customers order smaller and more often

Distributors increasingly prefer frequent, small orders over a month's stock at once. That pushes depots closer to markets, yet small, low-value parcels make freight a larger share of product cost. One planning head described the target plainly: very service-oriented and very cost-effective at the same time.

5. Limited visibility beyond the depot gate

Secondary and last-mile legs often run on third-party transporters without consistent tracking or digital proof of delivery. Without that data, it is hard to know how well each depot actually serves its catchment today.

Service levels hold when five things are planned together: lead distance, transport availability, inventory position, surviving-depot capacity and peak demand. Consolidation plans that address all five protect the delivery promise from day one.

Lead distance and transit time

Every merged depot extends some customer journeys. Network design maps each customer's latitude and longitude against candidate depots, so the new transit time to every pin code is known before the change, not discovered after it.

Transport availability in the new catchment

A depot location is only as good as the trucks that can serve it. As one buyer put it, the right site is one where transport is readily available and reach covers every market. That intelligence belongs in the model, alongside freight rates and vehicle types.

Inventory repositioned for the new flows

Stock that used to sit in five depots now needs to sit in three, in the right SKU mix. Safety stock, reorder points and transfer rules should be recalculated for the new footprint before the first depot closes.

Capacity at the depots that remain

Surviving depots absorb more volume. Docks, storage, loading time and operating hours set the real ceiling. Modeling these constraints prevents congestion and vehicle detention after go-live.

Peak and seasonal demand

A network that works on an average day can strain at month-end or in season. Testing each scenario against peak volumes is what makes the design robust.

The approach moves from data to decision to execution: build the baseline, fix the service target, find candidate locations, test scenarios, then redesign inventory and dispatch around the chosen footprint.

Step 1: Build one clean network baseline

Start with a single, trusted view of how the network runs today.

•     Demand: 12–24 months of order lines by customer, SKU and date

•     Locations: plants, depots and customers with accurate latitude and longitude

•     Capacity: storage, docks, loading time and operating hours per depot

•     Freight: primary and secondary rates by lane, vehicle type and mode

•     Constraints: vehicle restrictions, no-entry windows, product handling rules

Master data quality matters here. Vehicle dimensions, location masters and unloading times shape every result downstream.

Step 2: Set service level as a hard constraint

Agree the delivery promise before modeling begins: for example, 48-hour delivery to 95% of demand, or a defined service radius per channel. Encoding service as a constraint, not an outcome, means every scenario the model returns already protects customers.

Step 3: Find the right locations with center-of-gravity analysis

Center-of-gravity analysis weighs demand by volume and location to suggest where depots naturally belong. Layer on transport availability, highway access and labour markets to shortlist realistic candidate sites, including existing depots worth keeping.

Step 4: Model scenarios in a supply chain digital twin

A digital twin of the network lets planners ask “what if?” safely:

•     What if we merge these two depots into one?

•     What if this depot becomes a cross-dock instead of a stocking point?

•     What if month-end volume doubles in the north?

Each scenario returns cost-to-serve, transit time, service coverage, facility utilization and a robustness view, side by side. Trade-offs become transparent, and decisions rest on numbers everyone can see.

Step 5: Rebalance inventory for the new footprint

Once the footprint is set, inventory planning resets with it. Effective replenishment logic checks the nearest location with surplus stock before defaulting to the mother warehouse, recommends automatic stock transfers between depots, and adds buffer stock for the transit days a new lane introduces.

Step 6: Redesign dispatch and the last mile

Fewer depots means longer, denser routes. Dispatch planning should consolidate orders to the same distributor, choose the right vehicle mix, plan milk runs for nearby customers and respect time windows. Dynamic replanning then adapts plans when a truck runs late or an urgent order lands.

Track the same KPIs before, during and after rationalization, so every stakeholder sees the network improve on cost and stay steady on service.

Enmovil's AI-native platform connects network design, inventory and dispatch in one place, so the footprint decision and the operating plan are made together.

•     Network & capacity planning: center-of-gravity analysis, candidate depot and warehouse locations, and scenario comparison on cost-to-serve, service coverage and facility utilization.

•     Inventory planning and replenishment: SKU-level positions per depot, nearest-location transfers and automatic stock-transfer recommendations for the new footprint.

•     Dispatch planning: constraint-based plans with 50+ configurable constraints across vehicle type, time windows, pin codes and capacity, plus dynamic replanning.

•     Fleet sizing & allocation: the right dedicated vs. market vehicle mix for longer, consolidated routes.

•     Tracking & visibility and ePOD: real-time trip status and digital proof of delivery, so service levels are measured, not assumed.

•     CADDIE, the AI workforce layer: ask network questions in plain language and trigger planning actions from Microsoft Teams or Outlook.

The platform integrates with SAP and other ERPs, so the ERP stays the system of record while Enmovil adds the intelligence layer on top.

Depot rationalization delivers its full value when service level leads the design. With clean data, scenario modeling and inventory and dispatch plans built for the new footprint, manufacturers run leaner networks that keep every delivery promise.