Component-based software engineering (CBSE) is becoming the frequently
used approach for software development concerning time and cost
constraints. While most people’s lifestyle depends on the use of
software applications, the development cost and time of software
applications are becoming more difficult to achieve. Also, high-quality
and competent applications that fit users’ needs became essential.
Therefore, most software development organizations are reusing
commercial-off-the-shelf (COTS) to reduce development costs and time.
However, most organizations and developers face problems selecting
components that fit customer needs to integrate with the target system.
So, decisions regarding software component selection are hard to
consider regarding the entire quality of the software system. This study
aims to investigate the most important matters to software industry
practitioners and experts involved in component selection. First, the
author asked the practitioners to select the most important quality
criteria for an online bookstore from a list by providing their
subjective judgment and evaluation grades; after that, utilizing the
evidential reasoning (ER) approach for software component selection
problems due to its ability to analyze decisions with multilevel
evaluations and information uncertainty. Moreover, the main features of
the ER approach, such as weight normalization, probability assessment,
dealing with uncertainty, and utility intervals, offer several
advantages for COTS selection problems, including cost and time
minimizing, improved software reliability, effectiveness, and
efficiency. This study concentrated on assessing the quality criteria
for an online bookstore using the ER approach. Analysis results are
provided depending on the computational steps of this approach. Finally,
the findings show the rank of the four components according to their
weights, evaluation grades, and belief degrees for the selection.