---
title: "The walking is the cost: pick-path and slotting optimization for Dynamics 365 warehouses"
canonical_url: "https://cognilium.ai/blogs/warehouse-walking-is-the-cost-pick-path"
slug: "warehouse-walking-is-the-cost-pick-path"
section: "blogs"
date_published: "2026-10-12"
date_modified: "2026-10-12"
word_count: 665
reading_time_minutes: 3
author: "Mudassir Marwat"
author_identifier: "0009-0008-1927-2598"
person_same_as:
  - "https://www.linkedin.com/in/mudassir-marwat/"
  - "https://orcid.org/0009-0008-1927-2598"
  - "https://github.com/mudassirmarwat"
  - "https://medium.com/@mudassir-marwat"
  - "https://www.youtube.com/@mudassir-marwat"
  - "https://dev.to/mudassirmarwat"
  - "https://hashnode.com/@mudassirmarwat"
  - "https://huggingface.co/mudassirmarwat"
  - "https://bsky.app/profile/mudassir-marwat.bsky.social"
  - "https://substack.com/@mudassirmarwat"
  - "https://topmate.io/mudassirmarwat"
  - "https://fueler.io/mudassirmarwat"
  - "https://www.instagram.com/mudassirmarwat/"
  - "https://www.facebook.com/mudassir.marwat"
  - "https://www.reddit.com/user/mudassirmarwat/"
  - "https://www.f6s.com/member/mudassir-marwat"
  - "https://cognilium.ai/founder"
entities: []
related: []
---
# The walking is the cost: pick-path and slotting optimization for Dynamics 365 warehouses

Travel is roughly half of order-picking time, and your ERP records every move without ever optimizing it. Here's how pick-path and slotting optimization on Dynamics 365 cuts the walking.

Walk any warehouse at picking time and you will see the real cost of the operation, and it is not the picking.

It is the walking. Getting *to* the item — up and down aisles, back and forth across the floor — is roughly half of order-picking time in a typical operation. The actual grab-and-scan is the small part. Which means the biggest lever in the warehouse is not how fast people pick; it is how far they have to travel to do it.

Your warehouse manager almost certainly knows the pick paths are wrong and the fast-movers are slotted in the wrong places. What they usually do not have is a way to prove it, or a practical way to fix it — because the system that records every movement was never built to optimize it.

## Two problems hiding inside "picking is slow"

"Picking is slow" is really two separate decisions the ERP records but does not compute:

- **Slotting** — where each product lives. Fast-movers stranded at the far end of the warehouse, heavy items above light ones, products that are always ordered together stored aisles apart. Every one of those is a fixed cost paid on every single order until someone changes it.
- **Pick path** — the route a picker takes to fulfil an order. Even with perfect slotting, a bad route walks a picker past the same location three times or crosses the floor when a small reorder of the pick list would not.

Both are computable from data the ERP already holds — the item locations, the order history, what gets picked together, how often. And both are, in most warehouses, left to habit and the layout that happened to exist when the shelves went in.

## Your ERP records every move and optimizes none of it

Dynamics 365 warehouse management records the locations, the movements, the orders. It has the raw material for a much shorter walk. But it is built to *direct and record* the work — tell the picker where to go, log that they went — not to *derive* the arrangement and the route that would make the walk shortest.

So the slotting stays as it was, the routes stay as they are, and the travel cost gets paid, order after order, as an invisible tax on throughput.

## What a decision layer does instead

A decision layer sits on top of Dynamics 365 and does the warehouse math the ERP does not:

- **Slotting** — it computes where each product *should* live, from your own order history and pick-together patterns, so the items that move most are the ones people walk to least.
- **Pick path** — it sequences the pick work to cut travel, so a route stops doubling back.

The recommendations go back into the system for a person to approve and roll out — a slotting move planned into a quiet shift, a routing change the WMS then directs. As always:

- **It runs on your own history**, in your own environment.
- **A person approves the changes** — a slotting plan is not something you flip blindly.
- **It never touches the ERP core.** It reads and writes through the supported surface; your implementation and your partner are untouched.

Nobody picks faster. Everybody walks less. And because travel is the bigger half of the clock, less walking is where the throughput actually comes from.

## The same gap, across the operation

Pick-path and slotting are one of three places the same pattern shows up — the ERP records a judgement call it cannot compute. The others are [pricing](/dynamics-365/pricing-optimization) and [inventory and safety stock](/dynamics-365/demand-inventory-optimization). We build them as one family of [Dynamics 365 decision optimizers](/dynamics-365/optimizers), and the deeper method write-up lives at [pick-optimization methods](/dynamics-365/pick-optimization-methods).

If the walking is the cost in your warehouse and you want to see whether the fix is real for your operation, [tell us how your floor is laid out today](/contact) — you will get an honest read, including "your slotting is already good" when it is.

## Frequently asked questions

### What is pick-path optimization in Dynamics 365?

It is a decision layer that sequences pick work to cut travel and computes better product slotting from your own order history, then feeds the recommendations back into Dynamics 365 warehouse management for a person to approve.

### Why does travel matter more than picking speed?

Because getting to the item is roughly half of order-picking time in a typical operation. Shortening the walk moves the bigger number, so it does more for throughput than pushing pickers to move faster.

### Does the ERP not already handle this?

Dynamics 365 warehouse management directs and records the work, but it is not built to derive the shortest route or the optimal slotting. That calculation is what the decision layer adds.

### Will it reorganize our warehouse automatically?

No. It recommends slotting and routing changes; a person approves and schedules them. Slotting moves are planned, not flipped blindly.

### Does it change our ERP or our implementation?

No. It runs in your own environment on the supported surface, never alters core ERP logic, and complements your implementation partner.

---

Canonical HTML: https://cognilium.ai/blogs/warehouse-walking-is-the-cost-pick-path
