Remote instrument telemaintenance

Abstract

In the past decade, great technological progress has been made in telemaintenance of mainframe and mini computers. As hardware technology is now available at an acceptable cost, computer aided trouble-shooting can be adapted to laboratory instrumentation in order to significantly improve repair time, avoid instrument downtime by taking advantage of predictive methods, and provide general diagnostic assistance. Depending on the size of the instrument, the telemaintenance facility can be dedicated to a single instrument or alternatively a telemaintenance server can manage multiple distributed small instruments through a Local Area Network. As complex failures can occur, the local diagnosis capabilities may be exceeded and automatic dialling for connection to computerized Remote Maintenance Centers is needed. The main advantages of such a centre, as compared to local diagnosis systems, are the increased access to more information and experience of failures from instrument installations, and consequently the provision of training data updates for Artificial Neural Networks and Knowledge Based Systems in general. When an abnormal situation is detected or anticipated by a diagnosis module, an automatic alert is given to the user, local diagnosis is activated, and for simple solutions are given to the operator. In the last resort, a human expert can be alerted who, with remote control tools, can attend to the failures. For both local and remote trouble-shooting, the data provided by the instrument and connected workstation is of paramount importance for the efficiency and accuracy of the diagnosis. Equally, the importance of standardization of telemaintenance communication protocols is addressed.