MES System for Automotive: 8 Things to Check Before Implementation

An MES system for automotive manufacturing should let you quickly determine what happened to a product: which materials were used, which operations it went through, which parameters were recorded, and whether any quality issues occurred. This gives you a practical starting point for evaluating a system before implementation.

Our goal is to show you what to look for, how to test traceability, and what to ask an MES vendor before the project moves into the implementation stage.

If you are responsible for production or operations at an automotive plant, a list of MES features alone will not tell you much. Two systems may both offer OEE, reporting, traceability, and ERP integration, yet handle a material lot change, rework, a NOK result, or reconstruction of a recalled part’s history in completely different ways.

That is why it is best to start comparing solutions based on your actual production process and leave the module catalog for later, when it will be easier to make an informed decision.

Take the serial number or identifier of a product you manufacture. Try to determine what happened to it from the moment the material was issued until production was completed.

If answering that question requires opening your ERP, a local database at the machine, an Excel file, and a quality report, you will quickly see where information continuity is missing. An MES should connect these events with the appropriate production context.

The ISA-95 model places manufacturing operations management at Level 3 and describes its exchange of information with Level 4, which covers enterprise business functions.

On the shop floor, this creates a fairly clear division of responsibilities. ERP manages production orders and business processes, PLCs control machines, SCADA supervises the process, and MES connects production execution with the product, operation, and status.

A temperature reading of 183°C or an OK test result is not very useful without the right context. You also need to know which unit the measurement relates to, which workstation generated it, during which operation it was recorded, and whether the result fell within the required range.

In the automotive industry, traceability is directly connected with quality control. The IATF document covering minimum automotive quality management system requirements for sub-tier suppliers identifies product identification and traceability among the areas associated with automotive requirements.

A well-designed manufacturing traceability system should let you look at production from both directions. If there is a suspected issue with a particular material lot, you need to identify every product that used it. If you are investigating a complaint involving a specific part, you need to trace backward to the materials, operations, process parameters, and test results associated with it. Siemens describes these two directions as forward and backward traceability.

If your MES can locate a product but cannot identify all finished goods associated with a suspect material lot, your production traceability is incomplete and will not give you the results you need.

During an MES vendor presentation, it is easy to focus on the user interface, charts, and ready-made KPIs. A better test is to walk through events that actually occur in your production process.

To start, ask the vendor to show whether the system can answer the following questions:

Then reverse the scenario. Provide the lot number of a component and check whether the system can identify every product in which it was used.

A 15-minute demonstration like this may tell you more about an MES solution than a long list of available modules.

Consider a hypothetical automotive component production process. The ERP sends a production order, material is assigned to the process, and subsequent events are linked to the identifier of the part being manufactured.

At the first workstation, the machine performs an operation and sends relevant process parameters. The product then moves to testing, where the result is NOK. The system records the nonconformance and the product’s subsequent path.

The part is routed to rework and then tested again. This time, the result is correct. However, the previous NOK result, the corrective operation, and the new measurement should all remain visible in the product history.

This is an important characteristic of automotive manufacturing software. A current status of “OK” must not erase information about what happened earlier.

A good example is an explitia implementation for a manufacturer of foam components used in automotive seats, benches, and headrests. The MES system covered four production lines, connected planning with the current production status, and communicates bidirectionally with the line control systems.

Among other things, it records mold movements, foam dispensing, process parameters, downtime, and NOK causes while also sending production tasks to the line for execution. The project started with workshops involving the planning team, and the solution was launched line by line. This made it possible to validate the system against the actual manufacturing process.

This example shows that automotive manufacturing software should account for real-world limitations related to production lines, resources, and changeovers rather than simply record the completion of production orders.

In automotive IT environments, the boundaries between applications are not always obvious. The same report may sometimes be generated in MES, ERP, or a BI platform. Some process information may be available both in SCADA and in the manufacturing system.

Before implementation, you should therefore determine which system is the source of each piece of information and who is responsible for its accuracy.

This division may look different from plant to plant. ISA-95 structures functions and information flows between enterprise levels, but it does not impose one specific system or architecture.

For every important data field, determine where the information originates, where it should be stored, and which identifier will be used to connect it to the product.

This will help you avoid later disputes over system responsibilities and integration boundaries.

Assembly, machining, welding, and painting all require different things from an MES system. A universal checklist containing dozens of modules adds little value if half of them are irrelevant to your production line.

There are, however, several areas you should examine particularly carefully.

An MES implementation in the automotive industry is easier to plan when you know exactly what needs to improve.

The issue may be the time required to reconstruct product history, manual reporting, inconsistent quality data, or limited visibility into the causes of downtime.

Map the actual production process, including exceptions. Include rework, additional testing, manual decisions, and system behavior after communication is lost.

Situations like these should be included in your requirements from the beginning.

A data map will also be useful — and it does not have to be complicated. For every important piece of information, define its source, how it is acquired, how often it is updated, and which identifier will connect it to the product.

Only then can you have a meaningful discussion about integration scope and implementation costs.

MES is not a project owned by a single department.

Production understands how operations actually run. Quality defines traceability and product status requirements. Automation engineers know which data can be obtained from machines. IT is responsible for architecture, cybersecurity, and ongoing system maintenance.

Adding another system only makes sense if it organizes the flow of information between existing solutions. If it simply creates another isolated database, the problem may only move somewhere else.

At the final stage of comparing MES solutions, return to specific scenarios. The following questions will help you quickly identify areas that require deeper analysis:

You should also ask what is included as standard product functionality, what is configuration, and what requires custom development.

Features designed specifically for your process may be justified, but you need to understand their scope, how they will be updated, and who will be responsible for maintaining them in the future.

Launching the system on additional workstations does not, by itself, tell you whether an MES implementation has been successful. Before implementation begins, define a metric linked directly to the problem you want to solve.

If traceability is the priority, you can measure the time required to reconstruct a product’s history and the completeness of the information retrieved. If your objective is automated reporting, measure how much manual work was previously required to prepare those reports. If your goal is better downtime analysis, the quality and completeness of data about event causes will be especially important.

Before you start comparing vendor catalogs, choose one representative product and try to trace its path through your plant. See how quickly you can find the material, operations, process parameters, inspection results, and exceptions associated with it.

If you need several applications, spreadsheets, and help from multiple departments, you already have a strong starting point for a conversation about MES. The next logical step may be a joint workshop involving production, quality, automation, and IT, followed by preparation of a process and data map.

A good MES system for automotive manufacturing should quickly tell you exactly what happened to a specific part — and which other products could be affected by the same issue.

No. IATF 16949 defines quality management system requirements, including requirements related to identification and traceability, but it does not mandate the use of a specific MES product.

An MES system can support compliance with these requirements if it is aligned with the manufacturing process and customer requirements.

ERP primarily supports business and planning processes, while MES operates closer to production execution.

MES connects the production order with the actual sequence of operations and the information generated on the shop floor.

Yes, provided the required data can be obtained directly from the equipment or through an additional integration layer.

The amount of work involved depends on the PLC, available communication protocols, and the architecture of the specific production line.

No. An MES project can be implemented in phases as long as the target data and integration model is defined in advance.

The first phase can then be carried out in an area where the results are relatively easy to evaluate.