Railroads have been using wayside inspection technologies for many years to improve the safety and efficiency of operations. There has been a recent proliferation of new, more sophisticated, but also more costly inspection systems that are capable of detecting a wide range of subtle defects before a problem has actually occurred. This new class of “predictive” wayside detection systems, as compared with the older “reactive” technologies, requires a different deployment strategy. The higher cost of these new technologies means that it is particularly important for railroads to maximize the benefit they derive from the investment. This article presents a network optimization model that selects cost-effective installation sites for wayside defect detection systems over a railroad network. The objective is to maximize the total inspection benefits possible under any given investment budget. We develop solution techniques based on Lagrangian relaxation to effectively solve the problem. The article also presents case studies with empirical data to illustrate the technique. The computational results show that the problem can be solved efficiently, and that the model has the capability of being applied to full-scale railroad networks at regional or national levels. There are a variety of ways that railroads can use this model to help them more efficiently invest in wayside inspection technology so as to maximize the safety and economic benefits of these technologies.