Using k-means clustering in international location decision

Purpose
This paper aims to investigate the application of unsupervised machine learning in the international location decision (ILD). This paper addresses the need for a fast, quantitative and dynamic location decision framework.

Design/methodology/approach
Unsupervised machine learning technique, i.e. k-means clustering, is used to carry out the analysis. In total, 24 different indicators of 94 countries, categorized into five groups, have been used in the analysis. After the clustering, the clusters have been compared and scored to select the feasible countries.

Findings
A new framework is developed based on k-means clustering that can be used in ILD. This method provides a quantitative output without personal subjectivity. The indicators can be easily added or extracted based on the preferences of the decision-makers. Hence, it was found out that the unsupervised machine learning, i.e. k-means clustering, is a fast and flexible decision support framework that can be used in ILD.

Research limitations/implications
Limitations include the generality of selected indicators and clustering algorithm used. The use of other methods and parameters may lead to alternate results.

Originality/value
The framework developed through the research intends to assist the decision-makers in deciding on the facility locations. The framework can be used in international and national domains. It provides a quantitative, fast and flexible way to shortlist the potential locations. Other methods can also be used to further decide on the specific location.