Manufacturing Capability Assessment for Human-Robot Collaborative Disassembly Based on Multi-Data Fusion

Huiping Cheng, Wenjun Xu*, Qingsong Ai, Quan Liu, Zude Zhou, Duc Truong Pham

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)
365 Downloads (Pure)


In view of the fact that various resources are shared as services globally today in the manufacturing industry, the assessment and optimization for manufacturing capability of human-robot collaborative disassembly is the premise to realize the aggregation and optimization of the disassembly services, and provides the best basis for the optimal scheduling in the workshop. While human are the most basic manufacturing resource and industrial robots (IRs) are the most advanced, we establish a set of complete manufacturing capability assessment system and assessment model for human-robot collaborative disassembly in this paper. For the reason that most of the capability assessment method before ignored the data source selection of the assessment object, only used real-time data or historical data, this paper fuses the historical data and real-time data through manifold algorithm to get more accurate results. On this basis, we assess the manufacturing capability of human, robots, human-robot collaboration using the improved method combining PCA and Grey correlation degree method and AHP in disassembly process. Finally a case study is implemented to demonstrate the feasibility and effectiveness of the proposed method.

Original languageEnglish
Pages (from-to)26-36
Number of pages11
JournalProcedia Manufacturing
Early online date7 Jul 2017
Publication statusE-pub ahead of print - 7 Jul 2017


  • Assessment
  • Data fusion
  • Human-robot collaborative disassembly
  • Manufacturing capability

ASJC Scopus subject areas

  • Artificial Intelligence
  • Industrial and Manufacturing Engineering


Dive into the research topics of 'Manufacturing Capability Assessment for Human-Robot Collaborative Disassembly Based on Multi-Data Fusion'. Together they form a unique fingerprint.

Cite this