State of the art software development in the automotive industry and analysis upon applicability of software fault prediction / Harald Altinger.

Format:

Book

Author/Creator:

Altinger, Harald, author.

Series:

Audi Dissertationsreihe ; Band 19

Language:

English

Subjects (All):

Software measurement.

Computer software--Quality control.

Computer software.

Physical Description:

1 online resource (207 pages).

Edition:

1. Auflage.

Place of Publication:

Göttingen, [Germany] : Cuvillier Verlag, 2017.

Summary:

In recent years the amount of software within automobiles has increased up to 100 Million LOC in modern day premium vehicles. Virtually all innovations in automotive engineering in the last decade include software components. Parallel to this increasing amount, testing becomes more vital. Automotive software development follows restrictive guidelines in terms of coding standard, language limitations and processes. Traditionally testing is a core part of automotive development, but the raising number of features increases the time and money required to perform all tests. Repeating them multiple times due to programming errors might jeopardises a cars introduction on the market. SFP is a new approach to forecast bugs already at time of commit, thus to guide test engineers upon defining testing hotspots. This work reports on the first successful application using model driven and code generated automotive software as a case study and a success prediction rate up to 97% upon a bug or fault free commit. A compiled and published dataset is presented along with analysis upon the used software metrics. Performance data achieved using different machine learning algorithms is given. An indepth analysis upon factors preventing CPFP is conducted. Further usage and practical application areas will conclude the work.

Notes:

Includes bibliographical references.

Description based on online resource; title from PDF title page (ebrary, viewed September 27, 2017).

ISBN:

9783736985087

3736985088