<p>Acid fracturing is widely used to enhance the performance of oil and gas wells by improving reservoir permeability and maintaining pressure. This process is achieved by creating acid-etched fractures that facilitate the flow of oil and gas. In this study, a three-dimensional numerical model was developed, and the model parameters were evaluated using the Taguchi method. These parameters include permeability, Young’s modulus, closure stress, acid percentage, effective diffusion, viscosity, acid rate, and acid volume. The parameters were selected based on validation criteria comparing the 3D model, laboratory data, and field experiments. The model was validated through laboratory experiments on two core samples taken from the Sarvak (Bangestan) formation, the main reservoir in several southwestern Iranian oil fields. An artificial fracture was created along each core, and acid treatment was performed on the fracture surfaces for two durations—10 and 20 min—using a laboratory core holder apparatus. The maximum etching extent along the fracture geometry was compared between the experimental results and the 3D model predictions. The results showed the following errors: For the 20-minute tests: Core 1 (SABET- 1) showed an average error of 10.72% and a maximum error of 21%, while Core 2 (SABET- 2) showed an average error of 12.94% and a maximum error of 18.37%. For the 10-minute tests: Core 1 (SABET- 1) showed an average error of 14.02% and a maximum error of 36.46%, while Core 2 (SABET- 2) showed an average error of 12.59% and a maximum error of 19.83%. These findings indicate a reasonable agreement between the 3D model and the experimental results, confirming the model’s reliability in predicting fracture etching behavior. A computational approach was proposed for optimizing acid fracturing, using the conductivity of acid-etched fractures as a measure of effectiveness in relation to well production. Fracture geometry was initially determined using controllable design parameters within the GOHFER software. Additionally, fracture fluid characteristics, rock properties, and reservoir properties were incorporated into the model. The numerical framework, based on the SIMPLEM algorithm, iteratively calculates velocity and pressure components until convergence is achieved. Once these parameters are stabilized, the algorithm proceeds to calculate acid concentration at each node within the fracture area. The study found that formations with higher Young’s modulus produced better results with smaller acid volumes and higher flow rates, while formations with higher closure stress required larger acid volumes and lower flow rates. The outcomes of acid fracturing were strongly influenced by both formation characteristics and acid properties. To analyze the impact of various acid fracturing design parameters on fracture conductivity and geometry, the Taguchi method was employed, resulting in 27 experimental designs. By assessing these input parameters, the study aims to optimize the conditions under which acid fracturing can be most effective. This provides valuable insights into the design and implementation of acid treatments in reservoir engineering. Finally, the effect of various parameters on the dimensionless productivity index (JD) was investigated. Experiments 22, 17, and 24 had the highest JD values, with a peak of 0.32. Experiments 26, 20, and 13 exhibited the lowest JD values, at 0.49, 0.50, and 0.51, respectively. The 3D numerical model was validated using laboratory data from the Sarvak and Bangestan cores, obtained from carbonate reservoirs in several southwestern Iranian oil fields.</p>