The smart factory in automotive manufacturing
Just-in-time schedules and zero-defect expectations, from customers who can audit you. Where MES/MOM matters most for automotive suppliers.
Automotive has been doing lean manufacturing longer than most industries have been talking about it. The bar is high, and the customer sets much of it.
For tier one and tier two suppliers, the pressure comes from both sides. Customers expect just-in-time delivery and full traceability, with quality that's proven, not promised. At the same time, programmes change, volumes swing and the mix gets more complex every year.
MES/MOM is often already there in some form. The question is usually whether what's there can keep up.
Where the value usually is
Error-proofing at the station. Scanning the right part and revision, and checking the operator is qualified, before a job starts. In automotive, preventing the defect is worth far more than finding it.
Traceability at the right depth. Some components need serial-level genealogy, others lot level. Getting that decision right per part saves a lot of complexity. My comparison of lot and serial traceability goes through how to choose.
Quality data your customer will accept. Control plans, SPC on the characteristics that matter, first-article inspection, and records that stand up in an audit or a PPAP submission.
OEE you can trust. Most automotive sites measure OEE. Fewer measure it automatically, with the same definitions on every line. The difference is usually bigger than people expect.
What makes it harder here
Customer requirements are specific, and they differ by customer. Labelling and traceability rules can vary between two OEMs supplied from the same line. The system needs to handle that through configuration, not a custom build per customer.
Equipment is often a mix of generations. Connecting a new robot cell is easy. Connecting a press from the 1990s takes more thought, and it's usually the one you need data from.
And because many suppliers already have a homegrown or legacy MES, the project is often a replacement, which brings its own risks. I've written about replacing a homegrown MES.
Questions to settle before you look at vendors
- Which parts need serial genealogy, and which can be tracked by lot?
- Which customer requirements are we meeting manually today?
- How will we connect older equipment, and what data do we actually need from it?
- If we're replacing a legacy system, which of its quirks are real requirements and which are workarounds?
Common questions
Why do automotive suppliers use MES?
Automotive customers expect just-in-time delivery and quality they can audit. MES/MOM delivers those by error-proofing each station, recording genealogy at lot or serial level, running SPC on critical characteristics and capturing OEE automatically. Many suppliers already have some form of MES. The common challenge is whether it can scale across lines and sites.
Does every automotive part need serial traceability?
No. Safety-critical and high-value parts often need serial-level genealogy, while many others are well served by lot traceability. Deciding part by part keeps the system simpler and the data more useful. Your customer requirements will set the minimum for each part.
How does MES support PPAP and IATF 16949?
MES/MOM produces much of the evidence your quality management system relies on. Control plan checks executed and recorded at source, SPC on the characteristics that matter, first-article results and traceability records all come from the shopfloor. Pulled together, they make audits and PPAP submissions far less painful.
Start with these articles
Your traceability probably isn't as good as you think
Batch-level tracking and backflushing in ERP is better than nothing, but far from what's possible. What full traceability looks like in a recall.
Lot vs batch vs serial traceability: how granular do you need to be?
Lot, batch, serial and time-based traceability explained: how granular you need to be, whether lot and batch differ, and how to choose per product.
SPC on the shopfloor: catch the drift before it becomes scrap
Statistical process control explained for operations leaders. Control charts, capability, the rules that matter, and why SPC belongs in MES/MOM at the line.
Replacing a homegrown MES: the people problem nobody plans for
Replacing a homegrown MES brings challenges few anticipate: the people who built it, and the urge to rebuild the old system feature by feature.
What is OEE? How to calculate overall equipment effectiveness
How to calculate OEE step by step with worked examples, and why the bottleneck is the machine that really counts when you measure a line.
OPC UA, MQTT and UNS in plain English
The shopfloor connectivity layer without the rabbit hole. What OPC UA, MQTT and a unified namespace are, how they relate to MES/MOM, and what to ask vendors.
Other industries
Stuck on something? Ask me.
A question about your project, or about something you've read here. I read every one, and the useful ones become articles or FAQ answers, with names and companies left out.
This is a peer conversation, not a sales call or a consultancy. I work for an MES vendor, so I won't recommend a product, mine included.
Get the next one in your inbox
New articles on shopfloor AI, smart factory strategy and the MES/MOM lessons I wish someone had told me before my first implementation.
- Vendor-neutral, practical guidance
- Written from real implementations, not brochures
- Shopfloor AI, smart manufacturing and MES/MOM
- No hype, no buzzword bingo