Streamlining materials in flexible manufacturing systems is the key factor to greatly reduce idle time costs and labor costs. It is achieved by building location-aware robots that can estimate its position using off-the-shelf hardware. The location-aware robot can determine its position inside a building using online measured Received Signal-Strength Intensity (RSSI) from different wireless Access Points (AP), and an offline database of previously measured RSSI. The accuracy of the calculated online position can greatly affect the preciseness of the material handling. This paper studies the impact of the location of the wireless access points and their quantity in a manufacturing facility on the localization error. An optimum number of wireless access points and their locations is determined using a Genetic Algorithm for the case study of the Egyptian textile manufacturing facility. The optimum wireless access point placement in the 17 500 m² manufacturing facility improves the accuracy of localization by 46% using only nine APs to reach 4.7 meter error and 87% using 35 APs to reach 1.1 meter error, compared to the original plan of the wireless access points locations.