This study aims to identify the causes of delays in public construction projects in Thailand, a developing country. Increasing construction durations lead to higher costs, making it essential to pinpoint the causes of these delays. The research analyzed 30 public construction projects that encountered delays. Delay causes were categorized into four groups: contractor-related, client-related, supervisor-related, and external factors. A questionnaire was used to survey these causes, and the Relative Importance Index (RII) method was employed to prioritize them. The findings revealed that the primary cause of delays was contractor-related financial issues, such as cash flow problems, with an RII of 0.777 and a weighted value of 84.44%. The second most significant cause was labor issues, such as a shortage of workers during the harvest season or festivals, with an RII of 0.773. Additionally, various algorithms were used to compare the Relative Importance Index (RII) and four machine learning methods: Decision Tree (DT), Deep Learning, Neural Network, and Naïve Bayes. The Deep Learning model proved to be the most effective baseline model, achieving a 90.79% accuracy rate in identifying contractor-related financial issues as a cause of construction delays. This was followed by the Neural Network model, which had an accuracy rate of 90.26%. The Decision Tree model had an accuracy rate of 85.26%. The RII values ranged from 68.68% for the Naïve Bayes model to 77.70% for the highest RII model. The research results indicate that contractor financial liquidity and costs significantly impact construction operations, which public agencies must consider. Additionally, the availability of contractor labor is crucial for the continuity of projects. The accuracy and reliability of the data obtained using advanced data mining techniques demonstrate the effectiveness of these results. This can be efficiently utilized by stakeholders involved in construction projects in Thailand to enhance construction project management.

Noise pollution in construction sites is a significant concern, impacting worker health, safety, communication, and productivity. The current study aims to assess the paramount consequences of ambient noise pollution on construction activities and workers’ productivity in Peshawar, Pakistan. Noise measurements have been recorded at four different construction sites in Peshawar at different times of the day. Statistical analysis and Relative Importance Index (RII) are employed to evaluate the data Risk variables, such as equipment maintenance, noise control, increased workload, material handling challenges, quality control issues, and client satisfaction. The results indicated that noise levels often exceeded permissible limits, particularly in the afternoon, posing significant worker risks. In addition, RII analysis identified communication difficulties, safety hazards, and decreased productivity as significant issues. The results show that noise pollution is directly linked with safety risks, decreased performance, and client dissatisfaction and needs immediate attention by authorities. This paper proposes a strategic policy framework, recommending uniform hand signals and visual communication methods without noise for workers, worker training about safety, and using wearable devices in noisy settings. Communication training for teams and crane operators, proactive quality control, and customer-oriented project schedules are also proposed. These recommendations aim to mitigate the adverse effects of noise pollution, enhance construction industry resilience, and improve overall operational efficiency, worker safety, and client satisfaction in the construction sector of Peshawar, aligning with policy and sustainable development objectives.