<p>The rapid expansion of digital art markets has highlighted challenges in ensuring efficient, fair, and profitable auctions for non-fungible tokens (NFTs). Existing NFT auction platforms often face issues with optimizing auction parameters to maximize revenue while maintaining transparency and security. This work addresses these challenges by proposing a blockchain-based NFT auction framework that leverages an English auction model optimized through a Modified Dynamic Time-Extension Queueing Model (MDTEQM). To account for uncertainty, Monte Carlo simulations are used to evaluate auction performance, estimate expected revenue, assess the precision of these estimates, and verify the effectiveness and stability of the MDTEQM approach. Smart contracts following the ERC-721 standard were developed in Solidity, with ANKR facilitating transaction processing and IPFS via Pinata managing decentralized digital asset storage. Auction processes are algorithmically defined within smart contracts to optimize parameters for revenue maximization. Cost analysis demonstrates the solution’s economic feasibility. Performance evaluation using Hyperledger Caliper assessed latency, throughput, CPU, and memory usage across key token operations—createToken, buyToken, and resellToken. createToken exhibited the highest latency (up to 11.95 s) and lowest throughput, while resellToken showed the best performance with latency as low as 8.51 s and highest throughput. CPU utilization ranged from 70–80%, with memory usage averaging 675–755 MB. Monte Carlo simulations modeled dynamic bid arrivals and time extensions, demonstrating that increasing simulation sizes reduces variability and narrows confidence intervals in expected revenue estimation. The expected revenue stabilizes near $450 with higher simulations, balancing computational cost and reliability. This research offers a scientifically rigorous and practically scalable solution for next-generation NFT auction ecosystems</p>