Abstract— Data mining applications are becoming increasingly important for the wide range of manufacturing processes. During daily manufacturing operations large amounts of data is generated. The abundance of data however, often impedes the ability to extract useful knowledge. In addition, the large amount of data stored in often unconnected databases makes it impractical to manually analyse for valuable decision-making information. New intelligent Data Mining tools and techniques are required which can intelligently analyse data and produce useful knowledge for manufacturing. This is an important issue with regard to the development of an advanced maintenance strategy. Maintenance optimization is critical for enhancing the productivity of assets within an organisation. Maintenance effectiveness depends on the quality, timeliness, accuracy and completeness of the information related to asset optimization based on which decisions are made. Recently developed Condition Monitoring Systems (CMS) generate and collect large amount of data during daily operations. These systems contain hundreds of attributes, which need to be simultaneously considered in order to accurately model the system's behaviour and provide operators and senior management with the necessary data required to ensure production levels are met. This paper will present an overview of the big data tools and techniques required to collect and analyse a range of data to support the development of an advanced maintenance strategy. The challenges of big data in maintenance including capturing, accessing, and processing information will be analysed. To achieve e-maintenance, how to integrate information and communication technologies into maintenance and the corresponding requirements and constraints will be identified. I. INTRODUCTION ffective use of leading edge Information and Communication Technologies (ICT) is seen as important, and possibly critical, to the future competitiveness of European Industry. In particular, manufacturing organisations are frequently characterised by high staff turnover, lack of knowledge and training, and a lack of appropriate asset management strategies. This has resulted in poor manufacturing efficiency and large amounts of waste. The implementation of structured maintenance methods has made possible the development of ICT Dr David Baglee is with the Institute for Automotive & Manufacturing Advanced Practice (AMAP), University of Sunderland, Sunderland, SR5 3XB, UK (e-mail: David.baglee@sunderland.ac.uk , Dr Salla Marttonen is with the School of Business and Management, Lappeenranta University of Technology, FIN-53851 Lappeenranta, Finland. (e-mail: salla.marttonen@lut.fi ). Professor Diego Galar is with the Division of Operations and Maintenance, University of Lulea, Sweden(e-mail: diego.galar@ltu.se) including software and hardware systems. The production and process industry are passing through a continuous transformation and improvement for the last couple of decades, due to the global competition coupled with advances in ICT. Manufacturing organisations are focusing more on big data collection and analyses to support e-business intelligence. The data should also be used to support other functions within the organisation which could impact asset management such as marketing and customer relations. The aim is to remain competitive and efficient by improving equipment performance and reliability by introducing an asset management strategy based upon accurate data collection and analyses tools and techniques. Maintenance effectiveness depends on the quality, timeliness, accuracy and completeness of information related to machine degradation state, based on which decisions are made. This translates into two key requirements: (i) preventing data overload, ability to differentiate and prioritize data (during collection as well as reporting) and (ii) to prevent, as far as possible, the occurrence of information islands. With the emergence of intelligent sensors to measure and monitor the state of health of the component and gradual implementation of ICT in organizations, conceptualization and implementation of e- maintenance is turning into a reality [1]. While e- maintenance has a number of benefits seamless integration of ICT into the industrial environment remains a challenge. A variety of techniques are available to enable the above goals. Different data mining techniques serve different purposes, each offering its own advantages and disadvantages. The most commonly used techniques can be categorized in the following groups: Statistical methods, Artificial Neural Networks, Decision Trees, Rule Induction, Case-Based Reasoning, Bayesian Belief Networks, and Genetic Algorithms and Evolutionary Programming. It is very critical to understand and address the requirements and constraints from the maintenance as well as the ICT standpoints in parallel in order to identify and understand which information is required and when. II. BIG DATA BENEFITS AND CHALLENGES FOR MAINTENANCE Big data is a revolutionary advanced methodology where big data sets which are collected at an unprecedented scale, are often complicated and difficult to process using traditional data processing tools such as relational and object-relational database management systems. Big data refers to the datasets that could not be perceived, acquired, managed, and The need for Big Data collection and analyses to support the development of an advanced maintenance strategy Dr David Baglee, Dr Salla Marttonen, and Professor Diego Galar E