With the increasing complexity of wireless networks, the task of optimizing routing protocols for efficient data transmission becomes more and more challenging. Traditional algorithms frequently struggle to find a balance between multiple conflicting objectives, resulting in a suboptimal solution. This thesis addresses the problem of achieving the Pareto Optimal Front in Multiobjective Reinforcement Learning (MORL)-based wireless routing. The primary question addressed here focuses on how the Pareto Optimal Front can be constructed for MORL-based wireless routing that optimizes both packet delivery rate (PDR) and delay concurrently. We proposed a new framework for MORL-based wireless routing that uses a custom simulation environment named “Wireless Route Lake,” inspired by the Frozen Lake environment. This environment tries to create a realistic simulation of dynamic wireless network conditions, enabling the agent to learn and adjust its routing strategies. The suggested routing algorithm was implemented, tested, and improved iteratively using the Design Science Approach. An in-depth examination of exploration rates, Q-values, and performance metrics reveals the strength of the MORL framework in effectively managing conflicting objectives. The simulations’ results show that the proposed algorithm was able to build the Pareto Optimal Front with good performance, achieving an 85% PDR while effectively handling delays. This thesis successfully develops a robust MORL-based routing algorithm that balances PDR and delay, obtaining the Pareto Optimal Front in wireless networks.