In pavement management systems, it is beneficial to consider the economies of scale stemming from the synchronization of repairs conducted on neighboring sections within a single work zone. However, finding the globally optimal solution of the repair and work zone policy for a large-scale pavement network along a long-term planning horizon can be computationally cumbersome. In this study, as a benchmark, we first propose an exact solution algorithm based on dynamic programming. Then we second propose a computationally feasible methodology, a time-invariant simplified rule, to determine desirable (near-optimal) policies. The proposed methodology is applied to two numerical studies: (i) Case 1 for a small-scale road pavement system to compare life cycle costs and computational times between the rule-based methodology and the exact solution algorithm, and (ii) Case 2 for a real-scale road pavement system to discuss the effectiveness of the rule-based methodology. In Case 1, the rule-based methodology derives a near-optimal solution with a significantly shorter computational time than the exact solution algorithm. Case 2 shows that the rule-based methodology can find a superior policy to the aggregation of the optimal solutions independently found for each of decomposed sub-systems in a feasible computational time. Through sensitivity analyses, we find that the repair and work zone policies should vary depending on the deterioration process, cost factors, and weight between agency and user costs for society's view or available budget for the agency's perspective.