Currently, Combined Economic Emission Dispatch (CEED) is implemented via meta-heuristic optimizers in power systems to fulfill load demand. The primary objective of CEED is to optimize the power output of available generation units, thereby minimizing fuel and emissions costs. However, traditional CEED is typically achieved under conditions of abundant fuel supply, which may not be available across all operating scenarios, fuel types, or power systems. Thus, this article proposes a novel, simple, and effective enhancement to CEED for power systems under fuel shortage scenarios, enabling effective handling of fuel availability uncertainty. This study introduces the concept of Dynamic Generation Capacity (DGC), which updates the individual generation limits based on the available fuel quantity. This approach reinforces the main objective of the CEED process: determining the optimal output power for each unit while adhering to specified generation limits. Four different optimizers are considered: Walrus Optimization Algorithm (WaOA), Artificial Gorilla Troops Optimization (AGTO), Particle Swarm Optimization (PSO), and Genetic Algorithm (GA) to improve solution quality, accelerate convergence, and provide robust and comprehensive validation for the proposed technique. The standard IEEE 30-bus system is used to compare the performance of optimizers under different operating scenarios. The results showed that WaOA offered the most promising performance compared to the other candidates. For example, in the 870 MW load scenario without losses, WaOA reduced the generation cost by 0.0012%, 0.0005%, and 0.1026%, and the standard deviation by 99.99% compared to AGTO, PSO, and GA, respectively. The results also confirmed the efficiency, reliability, and simplicity of the proposed CEED with DGC under conditions of fuel shortage. DGC achieved 97.63% power security versus 97.01% for the best traditional CEED, reduced standard deviation by 99.9% compared to both traditional CEED methods, and improved response speed by 0.8% and 14.7% over the two traditional CEED techniques, respectively.