According to a recent survey the direct maintenance costs per year in European industry amount to 450 billion € of which 70 billion € (!) are estimated to be wasted through ineffective maintenance. Moreover, the indirect costs caused by degraded product quality, reduced production efficiency, loss of customers etc. are at least of the same range of magnitude as direct costs. A remedy is to move from outdated maintenance practices to modern ones based on automated condition monitoring (CM), able to timely detect the onset of fault and localize the root-cause. Since the traffic of data and information processing go in electronic format such a maintenance approach is referred to as emaintenance.

Why are then so few online condition monitoring systems implemented in practice in spite of tremendous advances in key enabling technologies, including sensors, processors, communication means and information technologies?

The answer is not as easy as is the question. Namely, the area of condition monitoring and condition-based maintenance is at the crossroads between technology, information services and human resource management. A closer look at the market landscape reveals that the bulk of manufacturers of on-line CM systems have targeted equipment from high cost range in sectors like power plants, wind mills and oil refineries, where high implementation costs can be more easily compensated by incomes typical for those branches. The same solutions, if applied to the equipment from middle or lower cost range (typical for “ordinary” industrial sectors), would encounter problems. Namely such investments turn not to be economically justifiable due to poor return of investment rates, overwhelming implementation costs as well as high costs of ownership.

These issues challenged the development of a CM platform referred to as MEMS-PHM for condition monitoring, prognostics and health management of industrial asset. It comprises a portfolio of MEMS sensors, smart nodes, communication interfaces, a MIMOSA database and reports generation modules (see Fig.1). The smart node (SN) is a low-cost, energy efficient and programmable unit, which can be installed on a broad range of assets distributed not only within a plant but also geographically. They perform periodical data acquisition and execute the dedicated signal processing algorithms used for diagnosis and prognosis. The results of processing are features that reflect the condition of the monitored machine. The results of processing as well as recorded data are transferred by various communication protocols to the central server within the company or via internet on remote server. A notable feature of the system is its ability to store the data in a way fully compliant with MIMOSA OSA-EAI standard. This open source environment enables integration with existing maintenance management systems, MES (Manufacturing Execution Systems) and ERP (Enterprise Resource Management) systems resident in the companies. Hence fully e-maintenance functionality is guaranteed to the end user. Minimization of the life-cycle costs is also due to a rich library of award winning algorithms, use of automatic code generation and novel self-tuning and learning capabilities.

Extending the MEMS-PHM portfolio with modules that allow for taking advantages of cloud technologies is obvious. Hence the cross-breeding of the technologies is conceived to additionally reduce the costs of CM solutions owing to access to data from anywhere and to anyone with approved access rights (developers, operators, maintenance people, management) without having to add additional expensive infrastructure.

There are some barriers that will have to be considered, e.g. company access policies and firewalls. Some companies will be reluctant to the open access to business data within the cloud/ web environment. The risk/ challenge will be to combine business critical and technical data to provide the advanced diagnosis and maintenance information required.