More Steel from the Same Machines
A leading steel manufacturer coordinating multi-stage production across geographically distributed plants.
better machine utilization from improved casting and rolling sequences
lower inventory carrying costs through better planning
planning cycles, with far less time spent on manual scheduling
The Challenge
Production spanned multiple interdependent stages, from casting to rolling mills, across geographically distributed plants. Planning ran on older, fragmented systems that limited visibility into core workflows, which made it hard to coordinate processes, manage inventory, and make decisions on complete information.
The operational complexity was severe: heterogeneous machines, non-uniform process flows, grade-mix constraints, frequent setup changes, irreversible production steps, and metallurgical requirements such as heat-wise batching and homogenization. Every stage constrained the others.
Planning was a weekly, manual exercise. The organization could not recalibrate quickly for inventory fluctuations, machine availability, or urgent customer requirements, and protecting high-margin products across facilities required constant manual alignment.
What We Built
We built a production planning engine that solves the full multi-stage, capacity-constrained orchestration problem: coordinating material flow, scheduling, batching, and sequencing across interdependent operations so that all customer orders are fulfilled in minimum total production time.
The engine optimizes at two levels. Monthly planning balances demand against available resources, allocates materials and machine time across facilities, and flags bottlenecks in advance. Daily operational planning then recalibrates production from real-time conditions such as inventory fluctuations, machine availability, and urgent customer orders, so the plan on the floor is always current.
On top sits an AI agent interface: planners work from a single dashboard and use natural language to test scenarios and analyze the impact of decisions before committing to them, with every answer grounded in the optimization models.
The Results
- Higher monthly production from smarter scheduling alone, without adding any new equipment
- 8% better machine utilization from improved casting and rolling sequences
- 12% lower inventory carrying costs, with shorter lead times and more reliable on-time delivery
- Fewer grade changeovers, cutting transition downtime between production runs
- Manual weekly planning replaced by optimized monthly plans plus daily operational replanning, freeing planners from most manual scheduling work
What changed: Production planning moved from a reactive weekly exercise to a two-level optimization capability: a monthly plan that commits the right orders to the right machines, and a daily plan that keeps it true as conditions on the floor change.
Next case study
Redesigning a Regional Supply Chain Network
What Would This Look Like in Your Operations?
Tell us about your operational challenge. We will give you a straight answer on what it needs.
Get in TouchThe first call is 30 minutes about your operations. No preparation needed, no obligation.