Efficient delivery scheduling remains a key challenge in transport logistics, especially under real-world constraints such as vehicle capacity, time windows, traffic conditions, and weather. To address this, a hybrid metaheuristic algorithm combining Adaptive Large Neighborhood Search (ALNS) and Tabu Search was developed, where an initial greedy solution is iteratively improved through global diversification and local optimization. The approach balances solution quality and computational time by integrating broad search mechanisms with focused refinements. A case study using real company data validates the method’s effectiveness in reducing route cost and improving operational efficiency. The results also highlight improved route structure and service consistency. This confirms the practical relevance of the model and its potential for broader application in logistics optimization.