Engineering Safer Commercial Vehicles: MSc Thesis in Safety and Systems Engineering

Thesis work is an excellent way to get closer to Scania and build relationships for the

future. Many of today's employees began their Scania career with their degree project.

Choose one of four thesis topics exploring how safety can be analysed, assured, and maintained in commercial vehicle operations. Topics range from AI-assisted incident analysis and modular safety cases to fallback strategies and motion planning under reduced braking capability.

Please indicate your preference regarding your chosen topic in the pre-screening question.

MSc Thesis Proposal 1

Agent-Assisted Safety Pattern Detection in Test Vehicle Operations

Background and Motivation

Semiannual reviews of test-vehicle incidents must combine records from different projects and reporting periods. Inconsistent categories, duplicates, and missing information make recurring safety problems difficult to identify. Repeating the analysis manually also consumes engineering time. An AI agent could execute a defined analysis workflow repeatedly, but its results must remain reproducible, traceable, and subject to engineer approval.

Objective

Develop and evaluate a reusable AI agent skill that prepares incident records, runs a defined pattern-detection workflow, and produces an evidence-linked draft report for each reporting period.

• Can an agent-executed workflow reduce preparation effort while preserving the correctness and traceability of a deterministic analysis pipeline?
• Which incident patterns add useful information beyond category counts, and how robust are they to missing, duplicated, or inconsistent records?
  
  

Job description

• Review incident-pattern methods and define a common CSV/Excel record structure using synthetic records representing two or three reporting periods.
• Implement a deterministic Python pipeline for schema validation, duplicate detection, missing-value flags, category counts, and one interpretable pattern-detection method.
• Define an executable AI agent skill with explicit inputs, tools, ordered steps, output schema, and stop conditions. The agent invokes the pipeline, flags ambiguous records, and drafts a report with links to supporting records.
• Add human approval for data corrections and safety conclusions; prohibit unsupported imputation, invented incidents, and autonomous release decisions. Log agent actions and preserve reproducible outputs.
• Demonstrate reuse on a previously unseen reporting period without modifying the agent skill. Compare agent execution with a manual workflow and the standalone deterministic pipeline; measure record-level accuracy, unsupported statements, traceability, run-to-run consistency, analyst time, required human intervention, and pattern stability under controlled data-quality changes.

Expected Outcomes

A versioned agent skill specification, reusable Python analysis code, synthetic test datasets, evidence-linked period reports, and a quantitative comparison. The contribution is an evaluated method for repeatable, human-supervised agent execution of safety analysis, not an autonomous safety decision maker.

Scope and Tools

One record format, two or three periods, one pattern method, and one agent workflow within 20 weeks.

No live company-system integration or confidential data is required. The student will discuss reporting bias and responsible use of AI in safety-critical engineering. Tools: Python, pandas, scikit-learn or mlxtend, CSV/Excel, and an accessible agent runtime; a scripted fallback remains usable without a commercial AI service.

Practical Information

Duration: 20 weeks. Location: KTH Royal Institute of Technology, Stockholm. 

Contact people:

Fatemeh Mohseni (Expert): [email protected], [email protected],

Patrik Jansson (Responsible manager): [email protected]

This MSc thesis will be funded by Scania.
The baseline project uses public or synthetic data and can be presented openly at KTH. Any industrial data use, publication arrangements, and student intellectual-property rights will be agreed before the thesis begins.

MSc Thesis Proposal 2

Modular Safety Cases for Autonomous Vehicle Updates

Background

A safety case explains why an autonomous function is considered acceptably safe by connecting claims, assumptions, and evidence. After an update, engineers must determine which parts of that argument remain valid. A modular structure may limit unnecessary review, but it can also hide important dependencies.

Objective

• Develop and evaluate a modular safety-case fragment for one autonomous stopping function, with explicit change-impact rules.
• How can a bounded safety argument be divided into modules without losing dependencies between claims, assumptions, and evidence?
• Does modularity reduce review effort without increasing missed change impacts compared with a conventional argument structure?
  
  

Job description

• Review modular assurance-case methods and select a public or literature-based autonomous stopping example with three to five claims.
• Represent the same argument as a conventional structure and as approximately three modules, recording explicit assumptions and dependencies.
• Implement a small Python demonstrator that identifies potentially affected claims and evidence for a specified change and explains each dependency path.
• Apply three bounded changes, such as a response-time revision, reduced braking capability, or replacement of verification evidence.
• Compare both representations against a reviewer-checked impact list, measuring missed dependencies, unnecessary review items, and review effort. Discuss how argument structure influences transparency and accountability.
  
  

Expected Outcomes

A reusable argument fragment, a dependency-aware change-review prototype, and evidence of when modularity helps or adds maintenance effort. Results will not imply certification or automatic safety approval.

Scope and Tools

One stopping function, three to five claims, and three changes within 20 weeks. No complete vehicle safety case or industrial access is required. Tools: Python, NetworkX, and a lightweight structured argument representation.

Practical Information

Duration: 20 weeks. Location: KTH Royal Institute of Technology, Stockholm.

Contact people:

Fatemeh Mohseni (Expert): [email protected], [email protected],

Patrik Jansson (Responsible manager): [email protected]

This MSc thesis will be funded by Scania.

The baseline project uses public or synthetic data and can be presented openly at KTH. Any industrial data use, publication arrangements, and student intellectual-property rights will be agreed before the thesis begins.

MSc Thesis Proposal 3

Deadline-Aware Fallback Planning for Autonomous Vehicles

Background

An autonomous vehicle motion planner may deliver a trajectory late or fail to deliver one at all. A fixed timeout is easy to implement, but the remaining time for a safe fallback depends on speed, obstacle distance, controller delay, and braking capability.

Objective

Develop and evaluate a supervisory switching rule that activates a predefined stopping fallback before the remaining stopping margin is exhausted.

• How can vehicle state, stopping capability, and bounded delays determine the latest acceptable fallback activation time?
• Can a state-dependent switching rule reduce unnecessary fallback interventions without increasing late responses compared with fixed-timeout and stopping-margin baselines?
  
  

Job description

• Review fallback supervision and select an existing runnable longitudinal planning and tracking-control baseline.
• Model one obstacle-approach scenario family with variable initial speed, obstacle distance, and delayed or missing planner outputs.
• Derive and implement a switching rule that includes stopping distance, control response delay, and bounded model uncertainty.
• Compare the rule with a fixed timeout and a simple stopping-margin or time-to-collision threshold under matched conditions.
• Measure stopping-margin violations, late switches, unnecessary interventions, computational cost, and sensitivity to uncertainty; identify cases where the rule is infeasible. Discuss risks of false intervention and missed fallback.
  
  

Expected Outcomes

A reusable supervisor prototype, reproducible simulation experiments, and a timing-budget method connecting planner delay, fault detection, and fallback response. No vehicle-level safety guarantee is claimed.

Scope and Tools

One scenario family, one stopping maneuver, and one planner-failure family within 20 weeks. Embedded implementation and vehicle certification are excluded. Tools: Python, NumPy, SciPy, and Matplotlib.

Practical Information

Duration: 20 weeks. Location: KTH Royal Institute of Technology, Stockholm.

Contact people:

Fatemeh Mohseni (Expert): [email protected], [email protected],

Patrik Jansson (Responsible manager): [email protected]

This MSc thesis will be funded by Scania.

The baseline project uses public or synthetic data and can be presented openly at KTH. Any industrial data use, publication arrangements, and student intellectual-property rights will be agreed before the thesis begins.

MSc Thesis Proposal 4

Adaptive Motion Planning with Reduced Braking Capability

Background

A trajectory planned for a fully functioning vehicle may become unsafe after braking capability decreases. A fixed conservative speed restriction can preserve margin but may unnecessarily limit operation. Fault detection and capability estimates may also arrive late or be uncertain.

Objective

Develop and evaluate a longitudinal motion planner that adapts speed and following-distance constraints to reported braking capability and its uncertainty.

• How should degraded braking capability, delayed fault reports, and estimation error be translated into planning constraints?
• Does capability-dependent planning improve stopping margin versus progress compared with unchanged and fixed-restriction planners?
  
  

Job description

• Review fault-tolerant planning and select an existing model predictive control baseline for two-vehicle longitudinal following.
• Model one bounded reduction in available braking deceleration, with externally supplied fault reports, detection delay, and estimation error.
• Implement capability-dependent speed and spacing constraints, including a conservative uncertainty bound.
• Compare the approach with an unchanged planner, a fixed conservative restriction, and a simple capability-aware rule under matched lead-vehicle braking scenarios.
• Evaluate distance-constraint violations, stopping margin, progress, braking comfort, solver feasibility, and computation time. Discuss the risks of incorrect capability estimates.
  
  

Expected Outcomes

A reusable MPC prototype and quantitative evidence on how braking degradation, reporting delay, and uncertainty affect planning performance. Findings may inform example interface requirements between fault monitoring and motion planning.

Scope and Tools

Two vehicles in one lane and one actuator-degradation family within 20 weeks. Fault-diagnosis design, detailed brake hardware, and real-vehicle tests are excluded. Tools: Python, CasADi, NumPy, and Matplotlib.

Practical Information

Duration: 20 weeks. Location: KTH Royal Institute of Technology, Stockholm.

Contact people:

Fatemeh Mohseni (Expert): [email protected], [email protected],

Patrik Jansson (Responsible manager): [email protected]

This MSc thesis will be funded by Scania.

The baseline project uses public or synthetic data and can be presented openly at KTH. Any industrial data use, publication arrangements, and student intellectual-property rights will be agreed before the thesis begins.

Education/program/focus: Master’s-level student in a relevant engineering or technology programme. Suitable backgrounds include electrical or electronics engineering, vehicle or automotive engineering, embedded systems, software engineering, computer science, information technology, mechatronics, robotics, automation and control, or systems engineering.

Number of students: 1

Start date for the thesis work: Flexible; to be agreed with the selected student. Jan/Feb 2027

Estimated time required: Approximately 20 weeks full-time (one semester), subject to the university’s thesis requirements and the agreed scope.

Application:

Your application must include a CV, personal letter and transcript of grades.

A background check might be conducted for this position. We are conducting interviews continuously and may close the recruitment earlier than the date specified.