From ad?hoc lists and calls to POS–WMS–TMS–driven, on?schedule store rebalancing

Overview

A fashion brand’s regional stock piled up in some stores while others ran out because transfers were coordinated with ad hoc lists, late-night calls, and manual carrier bookings. Intelligex connected point of sale (POS), the Warehouse Management System (WMS), and transportation through APIs, added an orchestration layer that proposed and triggered store-to-store transfers, and enforced approval gates and cutoffs. Inventory visibility sharpened, transfers fired on schedule, and the team stopped relying on emergency shipments and after-hours coordination.

Client Profile

• Industry: Fashion apparel and accessories retail
• Company size: Multi-region store network with central and regional distribution centers
• Stage: Established operations, modernizing inventory flow between stores
• Department owner: Procurement, Supply Chain & Logistics
• Other stakeholders: Store Operations, Merchandising/Allocation, Planning, Transportation, Finance, IT/Retail Systems, Regional Managers, 3PL Partners

The Challenge

Planners and regional managers tracked sell-through in spreadsheets and emailed lists of overstocked and understocked items to stores. Store teams confirmed availability by phone, printed labels, and asked transportation to arrange pickups. Transfers missed carrier cutoffs and arrived after promotions ended. Some stores over-shipped and created new gaps; others held inventory because they feared running short. The WMS and TMS were tuned for DC-to-store flows, not store-to-store rebalancing. As a result, stock sat in the wrong place while other locations issued rain checks or expedited replenishment.

The POS had near-real-time sales and on-hand data, but there was no shared logic to decide what to move, when to move it, and how to book transport. Allocation set rules in planning tools, but execution lived in emails and calls. Transportation booked couriers and LTL as one-offs with inconsistent carton specs and labels. Finance struggled to explain freight spikes tied to emergency transfers. Everyone agreed the brand needed to move inventory earlier and more predictably without replacing core systems or asking store teams to learn a new platform.

Why It Was Happening

Signals were fragmented. POS surfaced demand, the WMS knew what was locatable, and the TMS held carrier capacity and cutoffs. None of these systems enforced a single, multi-location policy for safety stocks, sell-through triggers, or lane selection. Rebalancing decisions depended on whoever had time to reconcile data and make calls that day.

Execution was manual. Store-to-store moves used improvised carton labels and free-text shipment notes, which created receiving errors and lost cartons. Transfers were created with little standardization on carton IDs or manifests, and carrier pickups missed windows because requests arrived too late. Without a governed workflow, late adjustments and exceptions multiplied, and the same stores were contacted repeatedly for status updates.

The Solution

Intelligex implemented a rebalancing orchestration that evaluated POS and WMS inventory against target positions, proposed transfers, routed them for human review when required, and booked transportation automatically. The approach integrated with existing retail and logistics systems: POS provided demand and on-hand, the WMS created transfer orders and carton IDs, and the TMS rated and scheduled pickups. Policies for safety stock, size curves, and promotional priorities were codified. The system produced compliant labels and advance ship notices, and every transfer carried a clear audit trail from trigger to receipt.

• Data integration across POS, WMS, and Transportation Management System (TMS), with POS as the demand signal, WMS as inventory truth, and TMS for rating and booking. Example TMS reference: Oracle Transportation Management.
• Rebalancing policy engine that applied safety stock by store, size and color curves by region, markdown and promo calendars, and store eligibility rules.
• Proposed transfer generator that grouped items into viable cartons by location and service level, with cutoffs aligned to store operating hours and carrier windows.
• Human-in-the-loop approvals for high-visibility moves, exceptions outside normal thresholds, and end-of-day cutover, with role-based permissions for Allocation, Regional Managers, and Transportation.
• Automatic creation of transfer orders in the WMS, carton and pallet labeling, and advance ship notices with Serial Shipping Container Codes (SSCC). Background: GS1 SSCC.
• Transportation booking via TMS APIs for courier, parcel, and LTL, with service selection based on cost-to-serve, promised arrival, and pickup cutoff.
• Inventory event model aligned to GS1 EPCIS concepts so movement events were traceable from store pick to receiving.
• Exception handling for inventory discrepancies, missing sizes, and store blackout periods, with automated reallocation or cancellation as needed.
• Dashboards for planners and regional managers showing over/under positions, approved and pending transfers, in-transit status, and aging.
• Workflow orchestration using a production scheduler to sequence checks, bookings, and confirmations on a predictable cadence. Example reference: Apache Airflow.

Implementation

• Discovery: Mapped POS data availability, WMS transfer order flows, and TMS carrier cutoffs by lane. Collected store-level policies for safety stock, size mix, and promo timing. Reviewed current labels, manifests, and receiving practices to standardize carton identification.
• Design: Defined the rebalancing policy model and priorities: protect safety stock first, then address overstock tied to upcoming markdowns, then balance size curves. Designed the approval gates and thresholds for automatic versus manual triggers. Specified API contracts for transfer creation, label printing, and booking.
• Build: Implemented integrations to read POS on-hand and sell-through, query WMS availability, and create transfers with carton IDs and SSCC labels. Built the policy engine and scheduler to propose and sequence transfers. Integrated with the TMS to rate options and book pickups, and set up event capture for pick, ship, and receive milestones.
• Testing and QA: Ran the engine in shadow mode to compare proposals with historical manual transfers. Injected edge cases like inventory inaccuracies, store blackout windows, and missed pickups to validate fallbacks. Verified labels scanned at origin and destination, and that events stitched together in the audit trail.
• Rollout: Started with a limited region and a focused set of categories, keeping manual overrides available. Enabled auto-booking for daytime windows first, then extended to end-of-day cutoffs. Expanded to additional regions after planners and store ops confirmed behavior matched policy.
• Training and hand-off: Delivered role-based sessions for Allocation on policy tuning, for Regional Managers on approvals and exceptions, for store teams on pick/pack/label scanning, and for Transportation on interpreting bookings and service selections. Published quick references and set up a shared channel for early-cycle support.
• Human-in-the-loop review: Transfers outside normal thresholds, high-ticket items, and moves near promo start required approval. All approvals, overrides, and cancellations captured rationale and owner in the log.

Results

Stores began receiving the right inventory at the right time because rebalancing decisions were made against shared policy and current data, not improvised lists. Proposals landed before carrier cutoffs, labels scanned end to end, and receiving matched manifests. Regional teams saw rebalancing alongside on-hand and sell-through, so they could adjust priorities without restarting the process.

Phone calls and emergency shipments tapered. Transportation bookings aligned with service goals and cost-to-serve, while Allocation and Planning gained a clear view of in-transit stock. Finance could reconcile transfer freight to policy choices, and store teams focused on serving customers rather than negotiating carrier pickups. The audit trail covered every step, from trigger to receipt, making process reviews straightforward.

What Changed for the Team

• Before: Email lists and late calls; After: Policy-driven proposals with scheduled orchestration and clear cutoffs.
• Before: Unclear on-hand and eligibility; After: POS and WMS aligned with safety stock and promo-aware rules.
• Before: Ad hoc labels and receiving issues; After: SSCC-labeled cartons and ASNs that matched at receipt.
• Before: One-off carrier bookings; After: TMS-rated and scheduled pickups with service aligned to priority.
• Before: Constant status checks; After: Dashboards and event traces from pick to deliver.
• Before: Exceptions handled by whoever answered the phone; After: Approval gates with rationale and audit history.

Key Takeaways

• Connect POS demand and WMS availability to a single policy engine; let rules, not email, drive rebalancing.
• Use the systems you have—WMS for transfer orders and labels, TMS for booking—and add a thin orchestration layer to coordinate them.
• Protect safety stock and promo timing first, then address overstock; encode these priorities so decisions are repeatable.
• Standardize cartons and manifests with SSCC and ASNs so stores receive what was shipped without guesswork.
• Keep humans in the loop for exceptions and high-visibility moves, but automate the routine proposals and bookings.
• Roll out by region and category, run in shadow mode, and expand once scanning and booking behave predictably.

FAQ

What tools did this integrate with?
The orchestration read demand and on-hand from the POS, created transfer orders and labels in the WMS, and booked transport in the existing TMS. Labels and manifests used GS1 identifiers such as SSCC. The scheduler ran on a standard orchestration platform, and dashboards consumed events from the same pipeline. References: GS1 SSCC and Oracle Transportation Management.

How did you handle quality control and governance?
Policies for safety stock, size curves, and promo priorities were codified and versioned. Proposals outside thresholds required approval, and all overrides captured rationale. Carton labels and ASNs enforced scan points at pick and receipt. Event logs aligned to GS1 EPCIS concepts, so each movement was traceable. Data validations blocked transfers when on-hand or location accuracy fell below acceptable confidence, routing to manual review.

How did you roll this out without disruption?
We ran the engine in shadow mode to compare proposals with historical transfers, piloted in one region and a limited category set, and kept manual overrides available throughout. Auto-booking was introduced during daytime windows first, with end-of-day cutoffs added later. Core systems and store processes stayed intact; the new layer coordinated them.

What about inventory accuracy at stores?
The workflow checked recent cycle counts and shrink signals. When confidence in on-hand was low, the system proposed smaller test transfers or required a quick scan count before approval. Discrepancies at pick triggered automatic reallocation or cancellation with alerts to Allocation.

How were carriers selected and costs controlled?
The TMS rated available services against promised arrival, pickup cutoffs, and cost-to-serve. Service selection followed policy by lane and priority, and exceptions required approval. Freight charges for transfers were tagged to the move and visible in dashboards, so Finance could connect costs to policy decisions.