Sustainability-Aware Collaborative Service Composition and Recommendation Based on Multi-Attribute Correlations

Abstract

In the context of network collaborative manufacturing and green manufacturing, some researchers have begun to focus on collaboration-awareness or sustainability-awareness service composition separately. However, service collaboration and sustainable performance are interrelated in the service business implementation, so they should be considered simultaneously when establishing service composition to ensure efficient service delivery. In this study, a sustainability-aware collaborative service composition and service recommendation (SCSR) model is formulated, aiming to pursue maximum customer service quality, as well as the synergy effect and eco-efficiency of combined services. Given the interdependencies inherent in service attributes, DEMATEL (Decision Making Trial and Evaluation Laboratory) is presented to acquire the influential weights for aggregating multiple service attributes into the service decisions. To recommend suitable composite services for core enterprises, NSGA-II with two initialization methods and three neighborhood structures is first proposed to produce high-quality solution sets regarding service composition. Then TODIM (an acronym in Portuguese of Interactive and Multi-Criteria Decision Making) is utilized to evaluate the relative dominance of service composition solutions under the psychological behavior of decision-makers. Numerous experiments demonstrate the effectiveness of the improved NSGA-II in handing the SCSR issue. Furthermore, the necessity of considering attribute interactions and synergy effects in service composition is verified.

Keywords: collaboration, Sustainability, service selection, service composition, attribute correlation