.jpg)
Gandhinagar, July 21, 2026: The Indian monsoon lasts from June to September. The heavy rains make repairs on some roads unavoidable. But long before the first drop of rain, there is an invisible critical force that must be taken into consideration. Concrete slabs are used to build rigid or concrete pavements. They scatter the traffic loads over a greater area. They are able to sustain large traffic loads. Because of these advantages, they're ideal for the construction of roads in highways and city roads. These are also used in the construction of pavements in airports.
Concrete pavements and the environment that surrounds them form a system that is bound together. These pavements undergo daily and seasonal heating and cooling. These fluctuating temperatures create stress that is internal to the structure of the pavements. This phenomenon is compared to the behavior of a chocolate bar to explain the concept. Chocolate has a lower structural integrity when it is kept outside and at room temperature. The texture of chocolate also deteriorates and becomes grainy and crumbly after some time. The same phenomena occur in concrete pavements. Repeated changes in temperature result in the deterioration of the structure of concrete pavements, which decreases the overall lifespan of the pavements.
To solve this issue, Indian Institute of Technology Gandhinagar (IITGN) researchers used machine learning to create a framework for designing custom, climate-resilient, region-specific rigid pavements. As published in the American Society of Civil Engineers (ASCE) Journal of Transportation Engineering, Part B: Pavements, this study shows a new, smarter way to design pavements. The focus of this research rests on the fifth largest state in India, Gujarat, where the coastal region is humid, and the areas of the state which lie further in are hot during the summer. Since Gujarat also has the longest coastline in the country, it has many regions with a lot of differing temperatures and a lot of differing wind patterns.
Researchers state that the standards present today in the estimation of thermal stresses for rigid pavements in countries of India and Nepal use climatic data from 1974, which may no longer be relevant. Many large climatic zones, where the classification is such that the entire states of Gujarat and Rajasthan are encompassed, undermine local thermal variations and greatly limit the reliability of predicting the performance of pavements and their design.
Describing the reasons for the study, Dr. Sumit Nandi, the study's author and former Postdoctoral Fellow in the Department of Civil Engineering at IIT Gandhinagar, noted:
“We started with the idea that it would be unwise to use the same design suggestions to build the roads if rigid pavements in different regions of the state would be subjected to different levels of thermal stresses.”
Dr. Nandi, a Scientist with CSIR-Central Road Research Institute and an Assistant Professor at the Academy of Scientific and Innovative Research (AcSIR), said the first step was to subdivide the State of Gujarat into 126 land-based grid points, and then collect the hourly climate data from the Copernicus Climate Change Service (C3S) via the ERA5 data. This was done for two climate time periods, 1961-1991 and 1992-2022, to analyze the changes in the Regional Climate.
This weather data was then used to perform the thermal modeling of rigid pavements. For this, three different Slab Thicknesses (200 mm, 250 mm, and 300 mm) were used in simulations, and two different surface albedos (i.e., Conventional Pavements and Cooler Pavements) of reflective nature were also considered.
The team then used a machine learning approach to recognize locations with identical thermal behavior after simplification of the data and loss of a minimal amount of key information. In this sense, the approach was analogous to the suggestion of a streaming service, where the system recommends shows/movies based on the user viewing history and preference, to group regions with comparable thermal heating, cooling, and stress, and also thermal equilibrium of the pavements.
The resulting analysis pinpointed five thermal microzones across the state of Gujarat. The zones had a linear temperature differential of approximately 16.3 °C to 17.2 °C, illustrating the difference in thermal behavior across the state.
While commenting on the results, Dr. Sushobhan Sen, Assistant Professor, Civil Engineering, IIT Gandhinagar, and Head of Built Environment Lab, stated,
"These results are fascinating and show that Gujarat does not display any rigid pattern for the thermal behavior of rigid pavements. Our study indicates that a design that is successful in one region of the state may not be sufficient to address the climate of another region of the same state."
He stated that the application of the framework across India would facilitate the development of thermal design pavements for rigid, state-specific, and climate-focused thermal design charts given that they allow for more rational use of expendable construction materials and balance durability and safety. He also added that the subsequent studies should consider the construction materials, as the quality of construction materials influences the lifespan of the pavements.
While India is undergoing another monsoon season, the study emphasizes that rain is not the only factor that determines the lifespan of a road. The seasonal variations of hot and cold as well as the long-term changes in climate also impact the lifespan of concrete pavements. The study is in line with the aims of PM Gati Shakti and the Bharatmala Pariyojana, and focuses on improving the transport infrastructure of India in the context of road development. The researchers also expressed thanks to IIT Gandhinagar on the post-doctorate fellowship awarded to Dr. Sumit Nandi.
Click Here for More Institutional Activities