The automatic orchestration and choreography of service-oriented architecture and web services is the problem area addressed in this paper. A comprehensive literature survey helps to define a boundary between preceding research and this work. The work attempts to answer this question: How can web services be composed efficiently and effectively, utilizing the available descriptions of other web services, into a new service that fulfills some prescribed requirements? The process of composition exploits the web service descriptions that are typically syntactic-oriented; however, numerous research studies have been carried out to enhance the accessible descriptive information with a semantic flavor. The paper narrows down the broad problem area on the nonfunctional requirements to satisfy the quality of service (QoS) requirements. The authors use a two-pronged approach: they define a framework that uses ontologies and description logics, and then they define a matching degree along with a strongly coupled scoring scheme that can measure the alignment between the requirements and the actually available properties of web services.

The authors perform an experimental evaluation of the proposed approach. The dataset used for the assessment is the Web Service Challenge 2009 dataset that contains Web services description language (WSDL) files, Web ontology language (OWL) files, and Web service level agreement (WSLA) files. A metaheuristic algorithm, harmony search (HS), is selected for seeking the matching web services. Extensive tables are provided that display the parameters of the searching algorithm that can be tuned; in addition, the achieved results are shown as “fitness,” that is, the goodness of the matching between web services. Furthermore, the measured running time and average deviation are shown.

The paper is interesting because it combines the ontology and description logics-based approaches with an algorithm that has roots in computational intelligence. The methodology for demonstrating and analyzing the achieved results can be considered useful as well because it can be perceived as a scientific method (meta-methodology) for proving the validity of the proposals and exhibiting the results in an empirical and repeatable way.

Those researchers who work in disparate domains such as web services and microservices and those who look for effective and semantic-based methods for orchestration and choreography of services will find the paper interesting. Moreover, researchers in computer science who want to prove the correctness of their ideas through empirical assessment will find the paper valuable.