This study introduces the ProACT (Probabilistic Analysis and Countermeasures Tool) framework, which uses Bayesian Networks, Colored Petri Nets, and NuSMV (new symbolic model validator) to check and improve smart contracts in blockchain-based supply chains. The main aim is to make smart contracts safer and more reliable in the Agrochem and Fertilizers Supply Chain. The research focuses on the need for better security in blockchain applications. There is a gap in current methods that handle the unpredictable nature of smart contracts. This study seeks to fill this gap with a new approach. The ProACT framework combines different models. Bayesian Networks help understand dependencies and uncertainties. Colored Petri Nets show how contracts change over time. NuSMV checks the contracts formally to find issues. The study includes a detailed case study and tests. Results show that ProACT is better at finding problems and checking contracts quickly. It also reduces the impact of attacks on the system. The findings are important because they offer a new way to make smart contracts safer. This helps improve the overall security of blockchain-based supply chains.