<i>K</i>-means clustering combined with principal component analysis for material profiling in automotive supply chains

Abstract

At a time where available data is rapidly increasing in both volume and variety, descriptive Data Mining (DM) can be an important tool to support meaningful decision-making processes in dynamic Supply Chain (SC) contexts. Up until now, however, scarce attention has been given to the application of DM techniques in the field of inventory management. Here, we take advantage of descriptive DM to detect and grasp important patterns among several features that coexist in a real-world automotive electronics SC. Concretely, Principal Component Analysis (PCA) is employed to analyze and understand the interrelations between ten quantitative and dependent variables in a multi-item/multi-supplier environment. Afterwards, the principal component scores are characterized via a K-means clustering, allowing us to classify the samples into four clusters and to derive di↵erent profiles for the multiple inventory items. This work provides evidence that descriptive DM contributes to find interesting feature-patterns, resulting in the identification of important risk profiles that may e↵ectively leverage inventory management for superior performance.

A data‐driven approach to improve online consumer subscriptions by combining data visualization and machine learning methods

“…Finally, the scores of all three variables are consolidated into a score (FTStrategies, 2022). As result of some successful cases around the media industry, the RFV became a strategic key performance indicator (KPI) across media companies to measure readers' engagement (Goad, 2016;Wenzel et al, 2022;Zontek, 2018). However, other approaches are emerging.…”

“…• Recency, frequency, volume: in 2016, the Financial Times marketing team adapted the renowned recency, frequency, monetary (RFM) analysis (Stone, 1989) to the media sector by defining the new engagement metric recency, frequency, and volume (RFV) (Goad, 2016). RFM is a simple but effective method used in marketing to analyse consumer purchasing behaviors (Coussement et al, 2014;Shen, 2021).…”

“…Furthermore, to measure reader engagement, a wide range of metrics are proposed in the literature (Davoudi et al, 2019;Goad, 2016;Grinberg, 2018;Peterson & Carrabis, 2008;Seale, 2021).…”

“…Cluster analysis is an important ML modeling technique to discover natural grouping of the observed data to create consumer profiles (Gonçalves & Carvalho, 2021; Silva et al, 2018; Vinothini & Priya, 2018). The concept of segmentation, firstly introduced by Smith (1956), consists in dividing a heterogeneous group into smaller homogeneous sub‐groups in which consumers share something in common (Fu et al, 2017; Smith, 1956).…”

“…In clustering, two learning problems are addressed that are supervised, also known as classification, and unsupervised (Rajput & Singh, 2023), known as clustering. The first referred to labeled data, the second one to unlabeled data (Gonçalves & Carvalho, 2021). One of the most popular and efficient clustering methods is the k‐means algorithm (Bock, 2007; Koul & Philip, 2021; MacQueen, 1967; Vergani & Binaghi, 2018) which has been evolving since the first approaches proposed in the fifties of the twentieth century (Bock, 2007; MacQueen, 1967).…”