Efficiency Analysis of Three Maze-Solving Algorithms

Abstract

Maze generation and solving have long been studied as fundamental problems in computer science and artificial intelligence, not only because of their recreational and educational value but also due to their relevance in real-world applications such as robotics navigation, game design, and optimization tasks. This study reviews real-world application scenarios of maze generation and solving algorithms and summarizes selected research advances in these areas over the past two decades. The research also evaluates the time efficiency of three classic maze-solving algorithms, which are Greedy search, Dijkstra’s algorithm, and Floyd-Warshall algorithm across different data scales, revealing insights into their performance and applicability in various contexts. The algorithmic time complexity and empirical runtime are analyzed and fitted to investigate the alignment between theoretical complexity and practical performance. Overall, this work highlights the importance of connecting theoretical advances with empirical evaluation, providing guidance for both academic research and industrial practice in algorithm design.