Increasingly, Intelligent Systems (IS) techniques are being used to solve both complex problems and problems with uncertainty. They also can implement the operator‟s knowledge (experience) into the system. This research aims to evaluate the well-known manufacturing metrics: the Overall Equipment Effectiveness, and Overall Line Efficiency, using IS techniques. Existing methodology to measure Overall Equipment Effectiveness (OEE) is based on three main elements consisting availability, performance and quality. This traditional method of measuring OEE has proven to be effective for batch production systems, also for production systems with the same weight of losses; however this method has some flaws. First, each element‟s weight is not equivalent with other elements since their losses are different [1]. For instance the quality rate associates with qualitative losses whereas availability is composed of time collapse. And second, for some of the continuous production systems such as oil, gas and petrochemical industries, computational efforts seem to be inefficient for calculating the performance rate due to lack of single unit of product. Calculating Overall Line Efficiency (OLE) which is an aggregation of all machine‟s OEE is also very useful for monitoring trends. However, for a factory with several manufacturing lines and machines with different levels of importance (weight factor), a new technique is required to measure the efficiency. In this thesis in order to enhance the computational method, efficiency and the option of considering scenarios with uncertain inputs, three methods: measuring OEE using Mamdani Fuzzy Inference Systems; measuring OEE using Sugeno Fuzzy Inference Systems; and measuring OLE using Fuzzy Inference Systems and Artificial Neural Networks are proposed. Additionally, an inverse relationship in Artificial Neural Networks is being proposed to achieve a certain line efficiency maintaining the machines‟ OEE. The proposed methodologies to improve the OEE and OLE weakness are based on Intelligent Systems techniques such as Fuzzy Inference Systems, and Artificial Neural Networks. These techniques result in an effective way to measure OEE and OLE considering different weight of losses and also the difference in machine‟s weight. Moreover, they allow the operator‟s knowledge to take a part in the measurement using uncertain input and output with implementation of linguistic terms. This thesis describes the proposed methodologies in detail, and the functionality of these methods is tested and the results are thoroughly analyzed.