Sustainability and reusability of learning objects are important. In addition, reusability and effective use of learning object are provided by metadata. There are massive multimedia materials in learning object repository (LOR). Therefore, selection difficulty of appropriate learning object (LO) issue is emerged. In this context, effective and reliable method has to be found to select reliable and suitable LO. In addition, searching and using a learning object in learning object repository (LOR) may take too much time. Generally, this searching process is done through metadata from LOR. If the selection criteria do not exactly match metadata values, it may not possible to obtain the appropriate LO. Analytic Hierarchy Process (AHP) which is a multi-criteria decision making (MCDM) method for addressing complicated problems reduces the waste of time and improves the accuracy of decision making. For these reason, AHP can meet the requirements of LO selection. In this study, the SDUNESA LOR software has been developed for selection of suitable LO by using AHP. This web-based software stores, shares and also selects most appropriate LO in the SDUNESA LOR. AJAX, XML and SOA Web Services are used in this software which is especially developed for computer engineering education. Criteria of AHP are defined according to the computer education priorities. Obtained results show that AHP supported SDUNESA LOR software selects the most reliable and appropriate LO that meets defined criteria. (C) 2014 The Authors. Published by Elsevier Ltd.