Trajectory Clustering Of Segmented Field Operations Logistics Process

It has become a new research direction to use process mining technology to mine and analyze all kinds of data generated during shipbuilding. The classification of hull segments into different clusters according to some properties in advance can make the excavation results more specific. In this paper, the process examples in ship construction are divided into different clusters based on the cohesive hierarchical clustering algorithm. Firstly, this paper introduces the advantages and disadvantages of several main clustering algorithms, selects the clustering algorithm most suitable for segmented outfield logistics, establishes the mathematical model of segmented outfield logistics process trajectory clustering, and defines the appropriate feature vectors. Euclidean distance, Hamming distance, Jekard distance and cosine distance were used to calculate the similarity distance between process instances, and the concepts of chain, single chain and group average were added to select the appropriate similarity distance between clusters. An evaluation method of clustering results based on coutour coefficient is introducedwhich can effectively evaluate the clustering results of the logistics process trajectory during segmented remote operations. Finally, the feasibility and effectiveness of the proposed algorithm are verified by experiments.