<p>Crop pests, including insects, pathogens, and diseases, cause major agricultural losses, a challenge exacerbated by global warming. These impacts generate significant economic and social costs, particularly in developing countries where agriculture underpins food security and livelihoods. As pests are adapting to pesticides, sustainable approaches such as intercropping and ecological practices are now crucial for effective crop protection. Mathematical modeling plays a central role in understanding crop pest dynamics and evaluating control strategies within integrated pest management (IPM). This study provides a comprehensive, systematic, and critical review of mathematical modeling approaches for evaluating crop pest management strategies. A structured search of Google Scholar, PubMed, Web of Science, and Scopus yielded 2126 records, of which 133 peer-reviewed studies met the inclusion criteria in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Deterministic models dominated the literature (54.89%), followed by optimal control models (30.08%); hybrid, stochastic, agent-based, and other specialized models together accounted for the remaining 14.89%. Most studies originated from Africa and Asia. This review critically assesses modeling assumptions, mathematical structures, parameter estimation methods, and the degree of empirical validation. It identifies key strengths such as analytical clarity, tractability, and policy relevance, as well as recurring limitations, including oversimplified biological assumptions, weak integration of field data, and limited generalizability across agroecosystems. Models incorporating optimal control theory provided clearer guidance for intervention planning. Overall, this review synthesizes current methodological trends, identifies persistent gaps, and proposes directions for developing more realistic, data-driven, and decision-oriented models to strengthen IPM and promote sustainable agricultural practices.</p>