Abstract
It is vital to optimize the drilling trajectory to reduce the possibility of drilling accidents and boosting the efficiency. Previously, the wellbore trajectory was optimized using the true measured depth and well profile energy as objective functions without considering uncertainty between the actual and planned trajectories. Without an effective management of the uncertainty associated with trajectory planning, the drilling process becomes more complex. Prior techniques have some drawbacks; for example, they could not find isolated minima and have a slow convergence rate when dealing with high-dimensional problems. Consequently, a novel approach termed the "Modified Multi-Objective Cellular Spotted Hyena Optimizer" is proposed to address the aforesaid concerns. Following that, a mechanism for eliminating outliers has been developed and implemented in the sorting process to minimize uncertainty. The proposed algorithm outperformed the standard methods like cellular spotted hyena optimizer, spotted hyena optimizer, and cellular grey wolf optimizer in terms of non-dominated solution distribution, search capability, isolated minima reduction, and pareto optimal front. Numerous statistical analyses were undertaken to determine the statistical significance of the algorithm. The proposed algorithm achieved the lowest inverted generational distance, spacing metric, and error ratio, while achieving the highest maximum spread. Finally, an adaptive neighbourhood mechanism has been presented, which outperformed fixed neighbourhood topologies such as L5, L9, C9, C13, C21, and C25. Afterwards, the technique for order preference by similarity to ideal solution and linear programming technique for multidimensional analysis of preference were used to provide the best pareto optimal solution.