Indonesia, as a vast archipelago, encounters substantial geographical challenges in ensuring reliable electricity access, primarily in rural areas. The utilization of renewable energy through DC Microgrid technology has the potential to enhance electricity availability and reliability in 3T (frontier, outermost, and disadvantageous) areas or rural areas. A cluster-based DC microgrid system, incorporating photovoltaics (PV) and batteries, can operate independently and power sharing between clusters. The optimal power flow (OPF) solutions ensure the most secure operating point for reliable power distribution. The proposed algorithm minimizes distribution losses using Dijkstra’s Algorithm approach. This approach calculates the weight for each path, identifying the shortest path in networks representing cluster-based DC Microgrid systems. Each weight corresponds to the power distribution loss between clusters. The analysis also incorporates varying State of Charge (SoC) condition of batteries in get optimal power flow. The optimal power flow results are validated using MATLAB Simulink and experimental results from previous studies. A multi-cluster scenario is implemented using a modified 9 IEEE bus system, considering different SoC conditions and demonstrating the optimal power sharing strategy between clusters. The proposed algorithm is implemented using MATLAB software and serves as a valuable tool for guiding the planning and design of DC microgrid systems, offering high adaptability for varying power source requirements. Furthermore, the result show that this approach enhances power reliability by addressing the challenge arising from power source fluctuations and load variations.