This research paper presents an experimental investigation conducted on a low-volume granular road pavement thickness, focusing on natural subgrade BR (%) and axle load repetition (msu). The main objective is to identify the best estimation of low-volume granular road pavement thickness parameters that lead to maximum granular layer thickness obtained from IRC: 37, shell oil, and AUSTROAD. At the same time, an accurate prediction model should be developed. 56 experimental runs involved two main granular road pavement thickness parameters: natural subgrade CBR (%) and axle load repetition (msu). Then, using the experimental data as a basis, two different prediction models for the granular layer thickness acquired from IR were created: 37, shell oil, and AUSTROAD. These models were created using the Flower Pollination Algorithm (FPA) and Adaptive Neuro-Fuzzy Inference System (ANFIS). Important conclusions show that, in terms of predictive accuracy, the ANFIS-based anticipated model outperforms the FPA model. The created ANFIS models were then optimized using the Flower Pollination Algorithm (FPA) in the subsequent work phase, which resulted in additional improvements. The research identifies optimal values for natural subgrades, and axle load repetition, demonstrating their effectiveness in maximizing granular layer thickness obtained from IR: 37, shell oil, and AUSTROAD. Specifically, the optimal parameters for maximizing granular layer thickness based on IR: 37 were determined as 540.45 mm for natural subgrade 3.8%, and axle load repetition 1.9 msu, maximizing granular layer thickness based on shell oil was determined as 609.619 mm for natural subgrade 3.4%, and axle load repetition 2.1 msu and maximizing granular layer thickness based on AUSTROAD were determined as 500.36 mm for natural subgrade 4.2%, and axle load repetition 2 msu, respectively. Most importantly, fresh trials were used to validate the prediction models, confirming their accuracy and dependability. The unique hybrid of ANFIS and FPA used in this work is noteworthy since it is a combination that is not often explored in studies on burr reduction. Through better granular road pavement thickness control, this hybrid strategy has shown to be a significant advancement in precision production, offering businesses a possible path to increase productivity, save costs, and improve product quality.