Role of Artificial Intelligence in Software Quality Assurance

This article discusses the importance of achieving software quality to ensure the safety and security of maritime systems. Software controls the behaviour of many maritime and offshore systems, making it an essential component of such systems. Software defects have been the primary cause of several accidents and have contributed to many more. All the good work of mechanical engineers and marine architects can be undone by a simple programming error. Unfortunately, standards and risk analyses that address software currently are not part of standard industry practice. This article reviews the motivation for addressing software as part of maritime systems assurance and summarizes the standards and risk analyses available to the industry. Both international standards and classification rules are beginning to be adopted, as part of an increasing recognition of the role of software in all complex systems. This article introduces a framework for examination of software quality, including process product, operational, and customer perspectives. Important elements of this article were introduced in Card [1]. This article focuses on software process quality. Getting the software right the first time is much cheaper than testing and fixing it to reach acceptability. The cost of rework for a typical software development project ranges from 20% to 50% of total project costs. This article identifies some common software process reference models and explains how they help to achieve software quality. While, there is no specific national or international regulation addressing Software Quality Assurance (SQA) in the Maritime and Offshore industry, some recommendations and guidelines have emerged recently. Three classification societies (DNV GL, ABS, and LR) have published software process reference models (standards) as optional rules for offshore vessels. These standards are based on widely accepted ISO/IEC standards that have been endorsed by national standards bodies. Regulations and liability concerns will eventually force the maritime and offshore industry to accept greater requirements for software quality assurance, as has happened in other major industries. For now, if a customer wants quality, he/she must ask for it. These new optional standards make that easier.

Download Full-text

The role of inspection in software quality assurance

IEEE Transactions on Software Engineering ◽

10.1109/tse.2003.1223642 ◽

2003 ◽

Vol 29 (8) ◽

pp. 674-676 ◽

Cited By ~ 42

Author(s):

D.L. Parnas ◽

M. Lawford

Keyword(s):

Quality Assurance ◽

Software Quality ◽

Software Quality Assurance

Download Full-text

Defining the role of software quality assurance in a medical device company

[1993] Computer-Based Medical Systems-Proceedings of the Sixth Annual IEEE Symposium ◽

10.1109/cbms.1993.263006 ◽

2002 ◽

Cited By ~ 3

Author(s):

K.R. Linberg

Keyword(s):

Quality Assurance ◽

Medical Device ◽

Software Quality ◽

Software Quality Assurance ◽

Medical Device Company

Download Full-text

Matrix-based component dependence representation and its applications in software quality assurance

ACM SIGPLAN Notices ◽

10.1145/1107541.1107546 ◽

2005 ◽

Vol 40 (11) ◽

pp. 29-36 ◽

Cited By ~ 11

Author(s):

Bixin Li ◽

Ying Zhou ◽

Yancheng Wang ◽

Junhui Mo

Keyword(s):

Quality Assurance ◽

Software Quality ◽

Software Quality Assurance

Download Full-text

Benchmarking PRAISE-CANDU 1.0 With xLPR 1.0

Volume 2: Computer Technology and Bolted Joints ◽

10.1115/pvp2013-98010 ◽

2013 ◽

Cited By ~ 1

Author(s):

Min Wang ◽

Xinjian Duan ◽

Michael J. Kozluk

Keyword(s):

Quality Assurance ◽

Pilot Study ◽

Fracture Mechanics ◽

Software Quality ◽

Verification And Validation ◽

Quality Assurance Program ◽

Software Quality Assurance ◽

Full Compliance ◽

Low Probability ◽

State Of Art

A probabilistic fracture mechanics code, PRAISE-CANDU 1.0, has been developed under a software quality assurance program in full compliance with CSA N286.7-99, and was initially released in 2012 June. Extensive verification and validation has been performed on PRAISE-CANDU 1.0 for the purpose of software quality assurance. This paper presents the benchmarking performed between PRAISE-CANDU 1.0 and xLPR (eXtremely Low Probability of Rupture) version 1.0 using the cases from the xLPR pilot study. The xLPR code was developed in a configuration management and quality assured manner. Both codes adopted a state-of-art code architecture for the treatment of the uncertainties. Inputs to the PRAISE-CANDU were established as close as possible to those used in corresponding xLPR cases. Excellent agreement has been observed among the results obtained from the two PFM codes in spite of some differences between the codes. This benchmarking is considered to be an important element of the validation of PRAISE-CANDU.

Download Full-text