How John Cockerill cut piping takeoff time by 80-90% with Mirage
John Cockerill's estimating and method teams used to read piping, boilermaking and structural steel drawings by hand. Mirage deployed a tailored AI platform that extracts quantities automatically and generates separate MTO and BOM outputs in minutes.
Published . Updated .
About John Cockerill
John Cockerill is an international leader in industrial engineering and advanced equipment, active across energy, defense, industry and infrastructure. Its Morocco operations run engineering and fabrication projects spanning piping, boilermaking and structural steel, each starting from 2D engineering drawings that estimating and method teams have to read, measure and turn into buildable quantities.
That work sits on the critical path of every project: nothing gets quoted, ordered or fabricated until the drawings have been measured and reconciled into a bill of materials.
Where it started
Before Mirage, quantity takeoff on 2D drawings for a commercial estimate took 3 to 4 days per project, and up to 15 days once suppliers were brought into the final costing. Getting an exact procurement list for fabrication was heavier still: the method team had to rebuild a full 3D model of the structure first, a step that could take up to a month before a single bill of materials existed.
On top of that, the 2D drawing title blocks and nomenclatures teams were working from regularly had omissions or manual counting errors, and there was no shared source of truth between the estimating team and the method/fabrication team, no history of how a given quantity had actually been measured and validated.
Inside the platform
Mirage built three purpose-built modules, one per engineering discipline, all running on the same underlying platform with a human-in-the-loop validation interface on top.
Piping
The module reads isometric drawings (ISOs) and automatically detects and separates two different outputs from the same drawing: raw MTO quantities for purchasing and costing (bar stock, elbows, flanges, weight, paint surface area) and a detailed BOM for the method and fabrication teams, broken down drawing by drawing, distinguishing shop welds from field welds.
Boilermaking
The same algorithmic engine handles both storage tanks and pressure equipment (like condensate receivers), adapted to the higher pressures the latter run at. It applies the API 650 standard to automatically calculate branch and flange lengths on shell courses, the industry's platform-clearance rule (150 mm above platform level), and roof/floor deductions for flange elevation, then generates the MTO automatically: tubes, blind and open flanges, dip pipes, shop and field welds.
Structural steel
Paper nomenclatures on 2D structural drawings are often incomplete or wrong. The module detects and links profiles (IPE and others) and plates, including gusset-plate detail views, and counts bolting exactly, bolt, nut and two washers, by clamping thickness. It handles symmetric assemblies, elevation changes and half-profiles, then generates a procurement list with profiles rounded up to standard 12 m bars and cutting waste optimized.
One drawing, two outputs
Every module produces two distinct, ready-to-use outputs from the same source drawing instead of one generic quantity list: an MTO for purchasing and commercial costing, and a BOM for the method/fabrication team, each in its own Excel format, each traceable back to the exact drawing it came from.
Human in the loop
The platform reads the drawings and applies the engineering rules; John Cockerill's estimating and method engineers validate. Every quantity is reviewable against the drawing it came from through a fast validation interface, giving both teams, and everyone who relies on their numbers downstream, a single, traceable source of truth over time instead of a spreadsheet nobody can fully audit.
What changed
Piping is live in production: John Cockerill's estimating team uses it directly, and the AI-assisted takeoff that used to take 3 to 4 days of manual counting now takes hours to minutes. Boilermaking and structural steel are built on the same engine, with the business rules validated against John Cockerill's own API 650 requirements and drawing conventions, and are rolling out next.
Where it goes next
- Boilermaking module: full production rollout
- Structural steel module: full production rollout
- Extending MTO/BOM automation to further drawing types
How the same takeoff feeds a quote: RFQ and quote automation for manufacturers.
The same piping takeoff for EPC contractors: AI piping takeoff, and what information take-off (ITO) covers. The method behind an MTO and a spool BOM, step by step: piping takeoff from isometric drawings.
Weighing manual takeoff software against an agent: PlanSwift alternatives.
Questions about this deployment
A tailored AI platform that reads piping, boilermaking and structural steel drawings and extracts the quantities and the fabrication bill of materials that the estimating and method teams used to take off by hand. Three engineering modules run on one platform.
80 to 90% less time spent on piping takeoff, and a fabrication bill of materials produced ten times faster: what took about a month is now a matter of clicks, with engineers reviewing the output instead of measuring.
An industrial engineering group whose estimating and method teams read piping, boilermaking and structural steel drawings for every quote.
Full production rollout of the boilermaking and structural steel modules, and extending the MTO and BOM automation to further drawing types.
See what Mirage can run in your operations
Show us the workflow that costs you the most time
Leave your work email, then pick a time for a 30-minute call. We look at one of your workflows together, and you leave with a written plan to automate it. No commitment.