A higher number of highway construction projects are experiencing delays in their scheduled completion time, and this situation, in turn, leads to overrun in the schedule as well as cost. This paper mainly categorizes the causes of time overrun and provides a mathematical model for predicting the time overrun percentage, using highway projects in Northern India as a case study. A survey of randomly selected samples yielded responses from 72 contractors and 44 clients. The study depends on forty-eight factors that influence schedule delays in highway projects. Based upon the average medium fuzzy influencing value and the respective frequency index value, the time overrun causing factors were ranked and categorized under different time overrun risk categories, namely red zone, yellow zone, and green zone. The analysis of the received responses indicates complications in the land acquisition process, utility shifting, constructability under traffic, lack of project planning and design changes were the top five most critical time overrun causing factors in highway projects. The medium fuzzy influencing value and the frequency index value for these five factors were above 18% and 80%, respectively. These factors lie under the red zone and need close attention. Factor such as opposing the projects by locals also gained much importance in this study as compared to previous studies as it lies under the red zone. The prediction model developed through the application of multiple regression analysis can explain 71.9% of the variability in the data, which is an excellent indicator of the model’s expected performance.

越来越多的公路建设项目的计划完工时间有所延迟，而这种情况又导致进度和成本超支。本文主要对时间超限的原因进行分类，并以印度北部的公路项目为例，提供了预测时间超限百分比的数学模型。对随机抽取的样本进行的调查得到了72个承包商和44个客户的回应。该研究取决于影响高速公路项目进度延迟的四十八个因素。根据平均介质模糊影响值和相应的频率指标值，将时间超限的产生因素分类并归类为不同的时间超限风险类别，即红色区域，黄色区域和绿色区域。对收到的答复的分析表明，土地征用过程中的复杂性，公用事业转移，交通情况下的可施工性，缺乏项目计划和设计变更是高速公路项目中最关键的五个时间超限因素。这五个因素的中度模糊影响值和频率指数值分别高于18％和80％。这些因素位于红色区域下，需要密切注意。与以前的研究相比，诸如当地人反对项目之类的因素在该研究中也变得非常重要，因为它位于红色区域内。通过应用多元回归分析开发的预测模型可以解释数据中71.9％的可变性，这是该模型预期性能的出色指标。