Neuro-fuzzy refers to hybrids of artificial neural networks and fuzzy logic where neural networks can learn from data, but cannot be interpreted and whereas fuzzy systems consist of interpretable linguistic rules, but they cannot learn. Using learning algorithms from the domain of neural networks to create fuzzy systems from data and then developing learning algorithms ensures models that can learn both fuzzy sets, and fuzzy rules, and can also use prior knowledge. The application of neuro-fuzzy to IT Asset Management and IT Infrastructure Management creates self learning models that will enable companies to drive down TCO through better utilization of IT Assets. ITAM and Neuro-Fuzzy The first phase in the asset lifecycle is procurement. When an asset is procured it enters the asset management system and needs to be managed. Ideally, the procurement system should feed the asset management system the data of the new asset as soon as the purchase order is completed and the receiving organisation should acknowledge receipt in a way that confirms the asset in the asset management system and notifies the purchasing system so that payment may be made. The procurement phase can be supported by Neuro-fuzzy models in the assignment of request for quotes to vendors with tenacity to deliver on time given other factors such as project urgency etc, or applying fuzzy logic to vendor selection via Black Economic profiles and supporting smaller to Medium enterprises. When users places request for new equipment to initialise the procurement of assets, then Integrated Purchase Requests are passed to the ERP application so that Purchase requests can be generated within the ERP system if so required for SOX and other audit purposes. PR numbers from the ERP system can then be used to drive the procurement process. Asset stock levels are validated and if in stock and according to predefined rules as defined during the planning phase, automatic issuing might be done. If no stock is available or if specific authorisation is required based on item/role/ etc the flow of the request will be queued to the next process. Approval can be tracked throughout the complete lifecycle of the assets. Whether it is to approve a request, approve an order, approve return requests, approve update of different databases or simply view the current status. Full workflow triggers are defined for the management of approvals with full exception reporting, reminders etc. Non-conformance to the process would result in the process not moving to the next phase in the procurement cycle. Collaboration and negotiation features allows for business case definitions to motivate requests between requestor and approver. Declining reasons are also sent to requestors. The second phase of the asset lifecycle is Commissioning. When an asset is commissioned, the system should be updated with relevant data such as location, responsible party in the organisation, configuration, vendor, warranty, and any other data that will be useful in managing the asset. The location may be a physical location or simply a link to some other asset that contains the asset being deployed. For example, software or memory will most likely be tied to a chassis and be located wherever that chassis is located. Finding trends between smaller Configuration Items is invaluable for predicting preventative maintenance tasks, procurement inputs, vendor selection etc. Automatically calculating refresh cycle of equipment via normal Logistical Analysis on Configuration Items (such as FRACAS, MTTF, MTTR, MTBF) etc can ensure properly operational equipment and reduce costs in that the procurement model and Configuration Item selections may be adapted. The third phase of the asset management cycle is utilisation. Utilisation is not simply a static flag but could be periodically updated by operational software that measures asset usage so that valuable assets not being used may be redeployed. Neuro-fuzzy clustering models, pattern matching and enhanced prediction models can drive cost of ownership down in that future license fees and spending patterns may be changed. Analysis of installed software in terms of business and non-business applications can discover usage trends by users to enable companies to do charge back to users for private use of business equipment. Within rental models the billing engines can be driven through further enhancements of such models.