Association rule mining using chaotic gravitational search algorithm for discovering relations between manufacturing system capabilities and product features

Abstract

An effective data mining method to automatically extract association rules between manufacturing capabilities and product features from the available historical data is essential for efficient and cost-effective product development and production. This article proposes a chaotic gravitational search algorithm–based association rule mining method for discovering the hidden relationship between manufacturing system capabilities and product features. The extracted rules would be utilized to predict capability requirements of various machines for the new product with different features. We use two strategies to incorporate chaos into gravitational search algorithm: one strategy is to embed chaotic map functions into the gravitational constant of gravitational search algorithm; the other is to use sequences generated by chaotic maps to substitute random numbers for different parameters of gravitational search algorithm. In order to improve the applicability of chaotic gravitational search algorithm–based association rule mining, a novel overlapping measure indication is further proposed to eliminate those unuseful rules. The proposed method is relatively simple and easy to implement. The rules generated by chaotic gravitational search algorithm–based association rule mining are accurate, interesting, and comprehensible to the user. The performance comparison indicates that chaotic gravitational search algorithm–based association rule mining outperforms other regular methods (e.g. Apriori) for association rule mining. The experimental results illustrate that chaotic gravitational search algorithm–based association rule mining is capable of discovering important association rules between manufacturing system capabilities and product features. This will help support planners and engineers for the new product design and manufacturing.

Digital Engineering Methods in Practical Use during Mechatronic Design Processes

“…Other applications support the concept phase by providing suggestions for later manufacturing strategies [30], manufacturing capabilities [31,32], and assembly [33]. Additionally, product configurations [34,35] and product family management [36] have been explored as use-cases as well as concept analyses [37][38][39][40].…”

A Combined Approach for Customer Profiling in Video on Demand Services Using Clustering and Association Rule Mining

“…In business and elsewhere, clustering and association rule methods are commonly used to define patterns in decision-making based on the data mining models. Association rule mining was proposed by Agrawal et al in 1993 [44]. The Apriori algorithm generates candidate (k-size) patterns incrementally and recursively to calculate and combine the frequent (k-1 sized) patterns [45].…”

“…Lift is helpful because it takes into account support as well as the dataset. The association rule has the form of X → Y and the formulas of support, confidence and lift measurements are given below [44]:…”

Association rules mining between service demands and remanufacturing services

“…Mohamed Kashkoush et al establishes a new knowledge discovery model based on historical manufacturing data to extract useful associations between manufacturing and design domains (ElMaraghy and Kashkoush, 2015; Mohamed Kashkoush and Hoda ElMaraghy, 2017). The data mining method proposed by Kou (2019) discovers hidden relationships between manufacturing system capabilities and product characteristics from historical data. It is used to predict the capacity demands of various machines for new products with different characteristics.…”