Why combinatorial testing works: Analyzing minimal failure-causing schemas in logic expressions

Author

Ziyuan Wang ; Yuanchao Qi

Author_Institution

Sch. of Comput. Sci. & Technol., Nanjing Univ. of Posts & Telecommun., Nanjing, China

fYear

2015

fDate

13-17 April 2015

Firstpage

1

Lastpage

5

Abstract

Combinatorial testing technique has been applied on Boolean-specification testing. Previous results indicated that it performed well to detect faults in Boolean expressions. This paper aims to investigate the reason why combinatorial testing works in Boolean-specification testing, which was always omitted previously. We design experiment to get the minimal failure-causing schemas for 19129 faults, which could be classified into 10 fault classes, of 20 Boolean expressions that extracted from TCAS system. By analyzing the number and strengths of minimal failure-causing schemas for each fault, the low bound of fault-detecting probability of τ-way combinatorial test suite for Boolean-specification testing are calculated. Computational results could explain the effectiveness of combinatorial testing technique in Boolean-specification testing.

Keywords

combinatorial mathematics; probability; program testing; software fault tolerance; τ-way combinatorial test suite; Boolean-specification testing; TCAS system; combinatorial testing technique; fault-detecting probability; logic expressions; minimal failure-causing schemas; Arrays; Conferences; Input variables; Software; Software testing; Telecommunications; Combinatorial testing; boolean-specification testing; fault-detecting probability; minimal failure-causing schema;

fLanguage

English

Publisher

ieee

Conference_Titel

Software Testing, Verification and Validation Workshops (ICSTW), 2015 IEEE Eighth International Conference on

Conference_Location

Graz

Type

conf

DOI

10.1109/ICSTW.2015.7107440

Filename

7107440