Loading…

Personalized manufacturing service composition recommendation: combining combinatorial optimization and collaborative filtering

Owing to the rapid proliferation of service technologies in cross-enterprise manufacturing collaborations, manufacturing service composition (MSC) has attracted much attention from both academia and industries. However, the existing service composition is often constructed by the combination of off-...

Full description

Saved in:
Bibliographic Details
Published in:Journal of combinatorial optimization 2020-10, Vol.40 (3), p.733-756
Main Authors: Zhao, Shuangyao, Zhang, Qiang, Peng, Zhanglin, Lu, Xiaonong
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Owing to the rapid proliferation of service technologies in cross-enterprise manufacturing collaborations, manufacturing service composition (MSC) has attracted much attention from both academia and industries. However, the existing service composition is often constructed by the combination of off-line and on-line services, quality of service (QoS) attributes are not appropriate for satisfying the specific demands of MSC. Moreover, there are very few historical QoS invocations of manufacturing service, leading to difficulty in recommending appropriate service composition to a target user. In order to find the personalized MSC mode from a complex service network more accurately, we combine combinatorial optimization with collaborative filtering in this paper to figure out two questions: (1) how to construct a QoS description model of manufacturing service composition; (2) how to enhance the effectiveness of personalized QoS-aware service composition recommendations. First, the new QoS model of MSC is proposed by considering both traditional characteristics (e.g. availability, performance and reliability), variability of service composition and enterprise dimensional QoS attributes. Second, the service combination optimization is constructed based on combinatorial optimization method. Third, the collaborative filtering is employed to calculate the missing QoS values of the candidate manufacturing services. Finally, with both available objective functions and predicted QoS values, optimal service composition recommendation can be generated by using combinatorial optimization model with QoS constraints.
ISSN:1382-6905
1573-2886
DOI:10.1007/s10878-020-00613-0