Human Errors in Software Engineering
This collaborative research project (joint with Dr. Gursimran Walia at North Dakota State University) looks to improve software quality in a new way by assuming that human error is a key cause of software defects. Research from cognitive psychology is used to develop a deeper understanding of the human errors that occur during the software development process and to develop techniques that detect and prevent those errors early in the software development lifecycle. Early elimination of mistakes will improve software quality and reduce overall development cost. Through the application of human error research from psychology, this work will improve developers? ability to identify, classify, and eliminate software development errors and provide a solid structure and theoretical basis upon which to build.
To effectively use findings from human error research to improve software quality, this project has three primary objectives. The first objective is to develop and empirically validate a requirement error taxonomy. The inclusion of human error research will ensure that the taxonomy is well-structured. Empirical evaluation with developers will ensure the taxonomy is complete and useful. The second objective is to use the taxonomy to build and empirically evaluate error-based software development techniques. These detection and prevention techniques will operationalize the error information into a format usable by developers. The third objective is to develop tool support for the error-based techniques. The use of human error research in this project will provide a more in-depth understanding of the types of mistakes developers make during development. In addition to its impacts on software quality, this project will also provide a venue for software engineering researchers to interact with cognitive psychologists, producing more diverse PhD students.
Working together with a Cognitive Psychologist, we have been conducting analysis of the software engineering literature and the human error literature to identify cases in which software engineering researchers have based their work on solid theories of human error from Psychology. In addition, we conducted the First Workshop on the Application of Human Error Research to Improve Software Engineering.
- Carver, J. (PI) "Integrating Software Engineering and Human Error Models to Improve Software Quality." National Science Foundation Award 1421006, 8/14 - 7/17. $269,758.
- Walia, G. and Carver, J. "Using Error Abstraction and Classification to Improve Requirement Quality: Conclusions from a Family of Four Studies." Empirical Software Engineering. 18(4):625-658. DOI: 10.1007/s10664-012-9202-3. Aug. 2013. (bib)
Walia, G. and Carver, J. "A Systematic Literature Review to Identify and Classify Software Requirements Errors". Information and Software Technology. 51(7):1087-1109. July 2009. (bib)
- Anu, V., Walia, G., Hu, W., Carver, J., and Bradshaw, G. “Using A Cognitive Psychology Perspectie on Errors to Improve Requirements Quality: An Empirical Investigation.” The 27th International Symposium on Software Reliability Engineering. Oct. 23-27, 2016. Ottawa, Canada.
- Hu, W., Carver, J., Anu, V., Walia, G., and Bradshaw, G. "Detection of Requirement Errors and Faults via a Human Error Taxonomy: A Feasibility Study." The 10th International Symposium on Empirical Software Engineering and Measurement. Sept. 8-9, 2016. Ciudad Real, Spain.
- Anu, V., Walia, G., Hu, W., Carver, J., Bradshaw, G. "Effectiveness of Human Error Taxonomy during Requirements Inspection: An Empirical Investigation." The 28th International Conference on Software Engineering & Knowledge Engineering. July 1-3, 2016. San Francisco, CA.
Workshops and Other Publications
- Walia, G., Carver, J., and Bradshaw, G. "Workshop on Applications of Human Error Research to Improve Software Engineering (WAHESE 2015)". 37th International Conference on Software Engineering (Volume 2). May, 2015. Florence, Italy. p. 1019-1020.
Projects Frontpage    |
List of Publications
Last Updated on August 3, 2015 by Jeffrey Carver