30 credits – Predic software releases using machine learning and AI

Scania is currently undergoing an exciting transformation from a traditional truck manufacturer to a provider of complete sustainable transport solutions. IT is a crucial part of this transformation as Scania’s success is depending on flexible and efficient IT solutions that support current and future business requirements.

30 credits – Predic software releases using machine learning and AI Ingress: Thesis project at Scania is an excellent way of making contacts for your future working life. Many of our current employees started their careers with a thesis project.

Background

The Devtools group at Scania is responsible for providing suitable development tools to the software engineers in Scania for them to swiftly build and release software into the cloud, self driving vehicles etc. Currently the tool portfolio comprises well known development software such as Github Enterprise, Gitlab, Atlassian Jira, Microsoft TFS, Perforce, JFrog Artifactory and more.

In our software development tools we have large amounts of data related to software projects regarding our self driving vehicles, cloud infrastructure, machine learning and AI projects as well as traditional software development projects. Today estimating when a new feature or bug fix will be deployed or released is typically done in a manual fashion using e.g. “best guesses”. It is well known that estimating how much time and effort a software development task will take is very difficult. This makes it hard for the business to do prioritizations and to plan product releases, educations and marketing activities. By using machine learning on all the data we have in our software development tools it should be possible to better predict and estimate software releases.

Target

The purpose of the thesis project is to develop a machine learning prototype that can estimate when software changes will be deployed into production and can be released.

Assignment

Investigate how the data in Scania’s software development tools, e.g. Gitlab, Jira can be extracted and structured for the purpose. - Develop a machine learning prototype that with existing data can predict when software features and changes will be deployed - Evaluate and quality assure the estimations of the prototype

Education


• Working on a masters degree in software engineering, informatics, computer science, business analytics, data analytics.
• Experience in software development tools, data analytics, machine learning, AI, programming (e.g. python).
• Excellent written and oral English communication skills.

Application 

Enclose CV, personal letter and school-leaving certificate.

Number of students: 2

Start date: ongoing/flexible

Estimated time needed: 20 weeks per student

Contact persons and supervisors

Mats Skoglund, Business Developer, IUAT, 08-553 528 29, mats.skoglund@scania.com

 


Scania is a part of Traton Group and one of the world´s leading manufacturers of trucks and buses for heavy transport applications. Scania is also leading provider of industrial and marine engines. Service-related products account for a growing proportion of the company´s operations, assuring Scania customers of cost-effective transport solutions and maximum uptime. Scania also offers financial services.
Scania’s IT organisation is an integrated part of the company providing products and services to all business units worldwide. Based on deep Scania business knowledge and internal relations we balance in-house core competences and service production with services externally sourced and governed by us. We are around 1500 people and the majority is located in Sodertalje.