Developing an NLP-enhanced sustainability balanced scorecard for construction companies

Date

2024-9-05

Author

Aali, Atefeh

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

50
views

0
downloads

Cite This

Due to global challenges, construction companies have implemented sustainable practices and reported their contributions towards sustainable development through annual sustainability reports. Aligning sustainability initiatives with strategic objectives and continuously monitoring progress is crucial for informed decision making. However, the lack of a comprehensive framework to evaluate construction companies’ sustainability performance has been a significant obstacle. This research introduces the sustainability balanced scorecard (SBSC), an innovative framework for evaluating sustainability performance in construction industry. The SBSC enhances the traditional balanced scorecard (BSC) by expanding it according to the triple bottom line principles. The absence of widely accepted or standardized indicators makes it difficult to identify and integrate the SBSC category indicators. Hence, to populate these aspects, sustainability reports were used to analyze key sustainability topics in the construction sector. The lack of a standardized format in these reports necessitated integrating natural language processing (NLP) techniques to streamline and improve their analysis. This study utilizes BERTopic, an unsupervised machine learning algorithm, to perform topic modeling and uncover the most common corporate social responsibility (CSR) topics and practices discussed in the sustainability reports of top international contractors. The topics and their representative keywords were then used to create a strategy map as a foundation for an SBSC and define metrics to achieve sustainability in the construction sector. The suggested tool was then verified in a case study on a sample Turkish company. The research findings will assist companies in publishing more comprehensive sustainability reports and provide managers with a novel performance evaluation tool designed for the construction industry.

Subject Keywords

Sustainability balanced scorecard, Topic modeling, BERTopic, Sustainability report, Natural language processing

URI

https://hdl.handle.net/11511/111039

Collections

Graduate School of Natural and Applied Sciences, Thesis