Over the clustered wireless network systems, development in wireless technology has had a more substantial influence. Entities need to communicate with one another in order to create a sustainable ecosystem. Clustering methods help connect and organise the sensor nodes by load balancing and extending the network lifetime. Only now, various techniques have been developed for solving routing problems but have yet to focus on routing reliability with avoidance of data collision in real scenarios. This research is carried out for the reliability of routing by multi-objective optimization in static and dynamic environments through agent-based analysis with avoidance of data collision and depletion of energy. This study introduces a fuzzy-based multipath clustering technique that exhibits both static and dynamic clustering formation properties. The designated region starts the clustering process once the sensor nodes are ready to begin the data transmission procedure. The proposed technique works in two steps: a) fuzzy cluster head selection; and b) multi-objective agent-based multipath routing protocols for shortest route path discovery. The enhancement made in cluster creation and selection is the critical feature. A well-organized sensor ecosystem has lessened the negative impacts of network collision and energy exhaustion. The packet delivery ratio, communication overhead, and energy consumption are the performance metrics examined when simulating the specified protocol using the computer language NS2. The devised fuzzy-based multi-path routing (FC-MRP) clustering technique outperforms the AODV (Ad-hoc on-demand distance vector routing) protocol, according to the results. The average percentage of improvement concerning PDR, Throughput, end-to-end latency, Overhead, Energy utilised, Energy efficiency, Network lifetime, and PLR is found to be +2.53, +2.23, -18.58, -22.46, -17.95, +23.00, +4.11, -18.09 respectively.