We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Plastic to apparel: an analysis of sustainable purchasing intention using a machine learning ensemble
Summary
This study used machine learning methods—Random Forest and Artificial Neural Network—to analyze factors influencing Filipino consumers' intentions to purchase apparel made from recycled plastic, achieving 91–92% classification accuracy. Customer perceived value and perceived behavioral control emerged as the primary drivers of sustainable purchasing intention toward plastic-derived clothing.
The use of plastics has become a significant component in maintaining the convenience and suitability of modern lifestyles; however, a vast majority of the million tons of plastic manufactured each year ends up in landfills, contributing to plastic pollution. With this, the fashion industry has capitalized to create recycled products. Despite the proliferation and continued presence of recycled and upcycle products, there still is a significant gap in the sustainable purchasing behavior of consumers. This study aimed to identify, analyze, and forecast the variables influencing consumers’ behavioral intention toward purchasing apparel made from plastic. This paper established the Sustainability Theory of Planned Behavior model to determine the purchase intentions of Filipino customers while purchasing clothing made of recycled plastic. A total of 500 valid respondents were gathered to evaluate factors: Perceived Economic Concern, Perceived Environmental Concern, Perceived Authority Support, Subjective Norm, Attitude, Perceived Behavioral Control, Customer Perceived Value, and Behavioral Intention. To analyze the data, the study utilized machine learning methods, such as Random Forest Classifier (RFC) and Artificial Neural Network (ANN). Data preprocessing using feature selection and correlation analysis was conducted to validate the available data, performed data cleaning process, and data aggregation. Several iterative processes were employed to generate the optimum classification model—obtaining a 92% accuracy for RFC and 91% for ANN at 150 epochs under 30 hidden layer nodes. With low error rates, the findings revealed that customer perceived value and perceived behavioral control were the primary factors influencing consumers’ behavioral intentions toward purchasing sustainable clothing. This study emphasized the consideration of these factors when planning marketing strategies and initiatives to promote sustainable apparel.