Elektroautomatik AB is a leading player in the automation industry, specializing in technology and automation. We want you to have the opportunity to work on something you are passionate about, leading to a more rewarding experience. Our focus is on matching the right person with thesis propsal.

The idea:

Safety functions for a machine are specified in a spreadsheet, the blockschema, and implemented separately as a program in a failsafe PLC. Nothing checks automatically that the two agree. One person writes the blockschema, another writes the safety program. The spreadsheet is also used for several tests in the machine during commissioning which confirms machine behaviour is according to safety concept.

This thesis builds a tool that reads the implemented safety logic out of a Siemens TIA safety project, reads the specified safety logic out of the blockschema, puts both into the same model and reports the differences.

A second part of the work is deciding what that result is actually worth: what can be claimed about safety once such a check exists and what still has to be verified by a person.

Background

The blockschema lists:

• Every safety element (emergency stop, light curtain, interlocked guard, safe door lock, muting sensor, two hand control)

• How those elements are grouped into safety areas

• Which actuators (drives, contactors, valves, brakes) are brought to the safe state by which area

The safety program reproduces that table closely enough that an experienced engineer can read one and predict the other. That is what makes the correspondence checkable by machine in the first place.

One problem though is change. An extra door is added to an area during commissioning, or a drive moves from area 3 to area 4. The spreadsheet is updated and the program is not, or the other way round. The mismatch survives until validation testing, or it is not caught at all. Validation according to ISO 13849-2 is manual and has to be repeated after every modification, which means in practice it is repeated less thoroughly than it should be.

Research questions

1. How completely can the implemented safety logic be recovered from a Siemens failsafe project, and what is lost in the process?

2. Is there a representation expressive enough for both the informal specification and the implementation, and still simple enough to diff and explain?

3. Which classes of deviation between specification and implementation are detectable this way, and which are not?

4. What confidence gain can be claimed from an automated consistency check, given how the functional safety standards treat software tools?

5. (Stage 4) Can a language model produce useful and traceable flags on safety concept documents without becoming a source of unjustified confidence?

What you get out of it:

• Functional safety work on real machines instead of textbook examples

• TIA Portal Openness, XML processing, graph modelling, static analysis of a graphical programming language

• A working knowledge of ISO 13849, IEC 62061 and IEC 61508, including the parts on software tools that most engineers never read

• A thesis with a concrete artefact and a concrete evaluation, which is a good combination to defend

Who are you?

You are in the final year of an engineering degree program specializing in automation, production, programming, or a related field.

You are able to be present at our facilities in Gothenburg at least three days a week.

Furthermore, we are looking for someone with a strong interest in technology and a desire to learn. You are responsible and structured in your work, and you excel at interacting, organizing, and coordinating your tasks with your supervisor. We also value a positive attitude and strong collaboration and communication skills.

What we offer:

We can offer you a role that is both challenging and tailored to your needs, ensuring you enjoy the assignment. You will be assigned a supervisor with whom you will have regular check-ins for support and guidance.

We also value a strong sense of community and therefore provide an office space for you to use during your thesis work, keeping you close to our team.

About Elektroautomatik:

Elektroautomatik is Sweden’s largest automation integrator and a full-service provider in the field of automation. We offer customized automation solutions, AGVs, high-quality machinery, training, and service that streamline and ensure the quality of our customers' production and logistics. With expertise spanning from initial concept to finished production and a broad range of competencies, we are a comprehensive partner for sustainable and innovative solutions.

We kindly request that advertising sales representatives and recruitment agencies refrain from contacting us regarding this recruitment process.