Bachelor and Master Theses

Title: EFFICIENT FBD TEST GENERATION THROUGH PROGRAM SLICING
Subject: Computer Science
Level: Basic
Description: Program slicing has been used as a way to improve the efficiency of debugging and program analysis.
Nowadays, automated test generation is being used to create tests automatically by analysing the
program structure and suggesting necessary tests. This kind of techniques are prone to efficiency
problems when applied to large programs containing many lines of code. This is especially true for
software in industrial practice, such as the ones running on Programmable logic controllers (PLC)
that are used in a wide variety of tasks from toys to avionics and rail transport. Function Block
Diagram (FBD) is a language used for programming PLCs that has gained lots of attention in
industrial practice due to its graphical nature. These FBD programs are often used in real time
and safety critical systems, and these systems require extensive testing in order to prove that they
pass safety requirements of their respective domains.
In order to improve the efficiency of applying automated test generation on FBD programs we
propose two FBD program slicing algorithms based on the structure of these diagrams and how they
are programmed. These FBD program slicing algorithms are applied on several industrial programs
for PLCs in order to evaluate their feasibility and what effect slicing has on test generation. Six
FBD programs were selected out of 189 provided by Bombardier Transportation Sweden AB. The
six programs were sliced according to these FBD slicing algorithms and then evaluated in terms of
efficiency and applicability.
Our results suggest that slicing can be used to reduce the needed generation time and memory
usage required for automated test generation. However, our observations show that slicing can also
increase the number of test cases generated and how this test cases are merged into one test suite
for the entire program.
Student: Christian Hurtig chg13006@student.mdh.se
IDT supervisor: Eduard Paul Enoiu
eduard.paul.enoiu@mdh.se, +46-21-101624
Examinator: Adnan Causevic
Adnan Causevic
adnan.causevic@mdh.se, 021-101689

Rapport och bilagor

Size

Senaste uppdatering


  • Mälardalen University |
  • Box 883 |
  • 721 23 Västerås/Eskilstuna |
  • 021-101300, 016-153600 |
  • webmaster |
  • Latest update: 2017.10.14