This study aims to replace the current conventional chilled-water system with solar-assisted absorption cooling in a large pharmaceutical company in Amman, Jordan. Conventional heating, ventilation, and air conditioning (HVAC) systems were used to serve the existing 9000 m2 area of the pharmaceutical facilities. Two solar absorption chiller configurations were proposed and evaluated to investigate this replacement using TRNSYS® software. The auxiliary heater and storage tank were configured in series and parallel and evaluated based on actual data from the company. The effects of the evacuated tube collector (ETC) on the solar fraction, storage tank volume (STV), auxiliary heat required (Qaux), and heat loss (Qloss) were studied. The grasshopper optimization algorithm (GOA), a newly developed optimization technique, was used to optimize the storage tank volume with solar fraction (SF), Qaux, and average heat loss (kJ/h) on an hourly basis in parallel and series configurations. Mathematical equations involving the factor such as storage tank volume and Qloss responses in parallel and series configurations were developed using regression modeling (RM) over all experimental observations. The optimal factor values were then calculated using the GOA and regression models. The confirmation findings indicated that SF was maximized, and Qloss was effectively minimized. Under the 95 % confidence level with = 0.05, the p-value for all RMs was <0.05, indicating the significance of the RM. Furthermore, the RM matched the provided data well, with a low prediction error. The GOA indicated that the parallel system configuration was the superior connection that provided optimum results for SF, and Qaux at a storage tank of 2.8 m3. According to the GOA method, the optimal STV for a series layout was 2.0 m3 and for a parallel arrangement was 2.8 m3. In terms of SF, the parallel system configuration outperformed the series system, which increased by 12.34 %. The parallel system significantly decreases Qloss and Qaux by 4.04 % and 7.39 %, respectively. To compare the GOA with other meta-heuristic optimization approaches, conventional HVAC systems were optimized using particle swarm optimization (PSO). The results indicated that the GOA surpasses the PSO technique in terms of the required convergence speed and population size. Because the movement of agents depends on the positions of each agent inside the swarm, GOA differs from other optimization algorithms. The novelty of this study demonstrates that the GOA technique provides a higher probability of convergence with less computing effort than well-liked and popular optimization methodologies such as PSO for a variety of Qloss, Qaux, and SF issues. These results have significant reverberation for the potential of the GOA to minimize Qloss and Qaux, and maximize SF in the parallel and series configurations of the auxiliary heater.