Automotive · Technology
MES and OEE dashboards for an auto components plant, on the machines it already had
A manufacturing execution system for an auto components plant that reads cycle counts and downtime from existing CNC machines and presses, without replacing the controllers. Supervisors log reasons on a tablet at the line, and the plant head sees OEE by machine, shift and part on one dashboard.
- Client
- A mid sized auto components manufacturer near Pune supplying machined and pressed parts to two wheeler and passenger vehicle OEMs
- Region
- India
- Sponsor
- Plant head and manufacturing engineering manager
- Duration
- 10 weeks for the pilot cell, 24 weeks to cover the full plant, then a quarterly review
Engagement model
O2E: Output Oriented Engagement
The first output was one machining cell connected and reporting OEE that the plant agreed was correct; the rest of the shop floor followed as a second priced output once that was accepted.
How O2E worksThe team
- Solution strategist with shop floor experience1
- Industrial integration engineer1
- Backend engineer2
- Frontend engineer1
- QA engineer (shared)1
The challenge
Where things stood
- Production was recorded on paper hourly boards and keyed into Excel the next morning, so the plant only found out about a bad shift a day later.
- The machines were a mix of ages and makes: newer CNCs with Fanuc and Siemens controllers, older turning centres and mechanical presses with no network port at all.
- Downtime reasons were written as free text, so the same breakdown appeared under a dozen names and could not be analysed.
- OEM customers were asking for part traceability and evidence of process capability, which the plant was assembling by hand for each audit.
The approach
How we ran it
Weeks 1 to 3
Shop floor survey
Walked every line with the maintenance team and listed each machine's controller, available signals and network options. Agreed a standard list of downtime and rejection reasons with supervisors before writing any code.
Weeks 4 to 10
Pilot cell
Connected one machining cell using native controller protocols on the newer CNCs and small edge boxes reading cycle and spindle signals on the older ones. Built the line tablet for supervisors and checked the system's counts against the hourly boards for three weeks until they matched.
Weeks 11 to 20
Plant rollout
Extended to the remaining machining lines and the press shop. Added shift planning, part and die changeover tracking, rejection logging and batch level traceability for the parts the OEMs asked about.
Weeks 21 to 24
Dashboards and reviews
Built live OEE boards on the shop floor TVs, a shift report sent on WhatsApp to the plant head, and Power BI reports for the monthly management review, all reading from the same data.
What it looks like
Representative screens
Drawn to show the kind of system we delivered. Client screens and data stay private.
Machining shop, shift B, 12 Oct
OEE, shift to date
71%
Planned output
4,800 pcs
Actual output
3,926 pcs
Open downtime
2 machines
- VMC-04 Fanuc82
- VMC-07 Fanuc74
- TC-11 Siemens68
- TC-02 (edge box)51
- HMC-0179
What we delivered
Modules and features
- Machine connectivity through controller protocols and edge boxes on older machines
- Supervisor line tablet for downtime reasons, rejections and changeovers
- Live OEE by machine, line, shift and part number on shop floor screens
- Shift planning against customer schedules with planned versus actual output
- Batch traceability linking heat number, die, machine and operator
- Shift summary sent to the plant head on WhatsApp
- Power BI reports for monthly loss analysis and customer reviews
What changed
Outcomes
- Supervisors and the plant head see output and downtime during the shift
- Downtime is logged against an agreed list of reasons, so the biggest losses can be ranked and worked on
- The plant kept its existing machines; older ones were connected with small edge boxes
- Traceability for OEM audits comes straight from the system for each visit
Solutions
Industry and platforms
Looking back
Spending three weeks matching the system's counts against the hourly boards before going wider felt slow, but it is why supervisors trusted the numbers; the first time a dashboard disagrees with a supervisor and the supervisor is right, the screen gets ignored. The older turning centres took longer than planned, because two of them needed custom wiring to pick up a reliable cycle signal, and we should have budgeted for an electrician's time from the start. We also learnt that the downtime reason list has to stay short. The first version had several dozen reasons and supervisors defaulted to 'others'; cutting it to a short list made the data far more useful.
Start a conversation
Discuss a similar project
Tell us where your team is today and we will walk you through how a project like this would run for you under O2E. The first conversation is free. Delivery work is paid only after you accept it.
A specialist replies within one working day. The first conversation is free.