Sensor networks are effective tools for detecting intruders. However, the standard technique of placing sensors in a perimeter is not optimal. Using optimization techniques to determine sensor placement can improve the effectiveness of the sensor network. The optimization should take into account the environmental conditions and place sensors to take advantage of these conditions. Additionally, there are multiple objectives to consider in sensor placement, specifically the probability of detection and the time to detect. Genetic algorithms are capable of optimizing both objectives simultaneously, achieving the Pareto-optimal curve. This allows the designer of the network to specify a necessary value for one objective and get sensor placements that optimize the other objective. Compared to the standard perimeter configurations, the genetic algorithm networks perform significantly better with respect to both probability of detection and time to detect.