GB2404463A - Searching and analysis of industrial data - Google Patents

Abstract

A method of searching and viewing industrial data items stored in one or more remotely-distributed industrial data sources 314, 318, 322, 326, 330, 332, 336 is implemented by a host server 342 interrogating the data sources to identify data items which satisfy a specified search criteria and transmitting to a client workstation 358 data representative of those data items which satisfy the specified search criteria. A pictorial representation of the industrial equipment is displayed on a visual display screen, the pictorial representation comprising one or more elements, the positions of the elements being determined by the values of corresponding measured parameters recorded in corresponding data items; and the elements being animated in accordance with changes in the measured parameters with time.

Description

IMPROVED SYSTEMS FOR SEARCHING AND ANALYSIS OF INDUSTRIAL

DATA AND THE PROVISION OF INDUSTRIAL KNOWLEDGE

This invention relates to improved systems for searching and analysis of industrial data and the provision of industrial knowledge. It is particularly applicable, but in no way limited, for use with data generated by control and monitoring equipment, associated with industrial process lines such as those found in steelworks or paper mills, or associated with marine vessels.

Background to the Invention

Industrial process lines, such as those found in steelworks or paper mills, employ a large number of electronic control and monitoring devices. These devices are associated with individual pieces of machinery in the process line, or control or monitor the behaviour of the process line as a whole. For example, a roller in a hot strip mill may be provided with devices monitoring the speed of the roller, the size of the roller gap, the roller temperature and the pressure it exerts on the steel under production. Other devices monitor attributes of the steel product itself, and provide information about the quality of the product. Quality control regulations often require this information to be supplied to the customer along with the finished product.

In many cases, the control and monitoring devices are computerised or processor controlled, and such devices may be referred to as data gatherers. Other types of data gatherers may be added to support maintenance and diagnostic functions, or production management. The term "data gatherer" as used herein should be interpreted broadly, to include those devices within the plant control/management system which collect and store electronic data as part of their normal function, and also any additional data-gathering devices which may be added external to the control/management system, either permanently hard-wired to a part of the plant, permanently connected to the plant via a network, or temporary (e. g. portable notebook computers, portable data-loggers which may be attached to monitoring and control equipment to obtain data as and when required).

Each data gatherer may gather and store a large amount of data, which will be referred to as a "data source". These data broadly fall into two types: (i) data generated continuously or periodically that may be used to create an ongoing record of the performance of a device or the condition of the manufactured product, and analysed to reveal anomalies or determine trends; and (ii) data generated in response to an event, such as when a device trips or fails, a threshold value is exceeded, or a fault is detected. A process line may employ many of these data gatherers, and the machinery will generally run as near to continuously as possible.

Accordingly it will be readily appreciated that a very substantial amount of data may be generated, day in day out, from the process line as a whole. Data sources can also be separate from the process line, and may be, for example, a report written by shift workers, or any other source of data.

The output data from the data gatherers are generally recorded as data items for subsequent analysis. The term "data item" as used herein is intended to be interpreted broadly, to encompass data files and records, and other collections and pieces of data. Many thousands of data items may be generated over a period of weeks.

The data items gathered by the data gatherers are of great value in the event of a fault or problem on the process line. For example, faults that result in the process line being shut down are highly undesirable. When this happens, scrap product is often produced, and the lost production time when the process line is non-operational is costly. When such problems occur, skilled experts need to search through the data items gathered by the data gatherers to try to identify the cause of the problem and analyse it. Given the sheer number and geographical distribution of data items that may be of relevance, this searching and analysis can be extremely time consuming and labour intensive, and can prolong the time during which the plant is shut down.

Even when problems do not occur, it is desirable to be able to monitor and analyse the data being gathered by the data gatherers, to establish trends in the behaviour of the plant and to identify potential problems that may be likely to occur. There is benefit in proactively identifying a potential problem before it happens, and correcting it before damage and/or lost production time is incurred.

An associated issue under consideration is that of knowledge bases. These are sources of knowledge that are intended to be consulted when a problem occurs on the process line, to provide guidance as to how to overcome the problem. In an industrial context knowledge bases have inherent problems and issues that must be addressed. For example, who owns the intellectual property or industrial know-how associated with the knowledge? How and where should this knowledge be stored, to satisfy the owners of the intellectual property and industrial know-how, yet making it available to all the relevant users?

Summary of the Invention

According to a first aspect of the invention there is provided a method of searching and viewing industrial data items stored in one or more remotely-distributed industrial data sources, a data item comprising: industrial process-related data, identification data by which the data item can be identified, and searchable data representative of industrial process criteria; the method being implemented by a host server and comprising: establishing a data communications connection, via a network, between the server and a client workstation on which the data items are to be viewed, and between the server and said one or more remotelydistributed data sources; initiating a search of the data sources by processing data specifying industrial process or other criteria by which the data sources are to be searched; interrogating the said one or more data sources to identify data items which satisfy the specified search criteria; transmitting to the client workstation data representative of those data items which satisfy the specified search criteria; and enabling a user to view the content of one or more of those data items which satisfy the specified search criteria. This advantageously enables potentially a large number of data items from a plurality of different data sources to be searched, and for data to be readily obtained that is relevant to a problem, or of interest in monitoring a plant or establishing trends etc. The term "industrial process" as used herein should be interpreted broadly, to encompass all industrial or engineering applications, including factories, process lines, power generation, power transmission, power distribution, power conversion, and transportation applications (e.g. marine drives).

Preferably the step of initiating a search of the data sources by specifying industrial process or other criteria comprises receiving input from the user, via the client workstation and the network, specifying the desired criteria. Thus the user enters criteria by which the server performs a search of the data sources.

Alternatively, the step of initiating a search of the data sources by specifying industrial process or other criteria may comprise running an automated agent, which specifies pre- determined criteria and initiates pre-determined actions. Such an automated agent enables data to be obtained without user input at the time, either according to a pre-prepared schedule (e.g. weekly) or in response to pre-determined conditions having been met (e.g. an alarm having been activated on the process line).

Preferably the step of enabling a user to view the content of those data items which satisfy the specified criteria comprises: displaying on the workstation the identification data of those data items which satisfy the specified criteria; enabling the user to select a data item of interest; and displaying the content of the selected data item on the workstation. Thus the user may scan through a list of the data items that have been found and select one or more to view. The user may select a plurality of data items of interest, and those data items may be downloaded to the workstation in a packaged form such as a ZIP file.

The step of displaying the content of a selected data item may further comprise using a pre- prepared template, the content of the selected data item being displayed in accordance with conditions defined in the template.

Preferably the step of displaying the content of the selected data item comprises displaying the content graphically. This may be done using proprietary software on the client workstation, or third party software for graphing or analytical purposes. In a preferred embodiment the content of the selected data item may be displayed in an animated form to show changes in measured data over time.

The method may further comprise receiving input from the client workstation specifying data source types to be searched. Where necessary, a data source interface or driver may be received and installed, to enable a hitherto unrecognised type of data source to be recognised and searched. Software (referred to as a package handler) may also be installed to enable a hitherto unrecognised type of data item to be recognised and viewed either via a proprietary or a third-party graphic or analytic software package.

Input may be received from the client workstation specifying data sources or data items to be searched. Alternatively, or additionally, input may be received specifying one or more variables by which to search, the variables being selected from a group comprising: time, location, text content, industrial equipment identity, product-type.

According to a second aspect of the invention there is provided a host server configured to execute a method of searching and viewing industrial data items in accordance with the first aspect of the invention.

According to a third aspect of the invention there is provided a data communications network comprising one or more remotely-distributed industrial data sources and a server in accordance with the second aspect of the invention.

According to a fourth aspect of the invention there is provided an industrial plant or process line comprising a plurality of data gatherers configured to gather industrial process-related data, said data being stored in one or more remotely-distributed industrial data sources in communication with a data communications network in accordance with the third aspect of the invention.

According to a fifth aspect of the invention there is provided a computer program executable by processing means to cause a server to execute a method of searching and viewing industrial data items in accordance with the first aspect of the invention.

According to a sixth aspect of the invention there is provided a computer program in accordance with the fifth aspect of the invention, stored on a data carrier.

According to a seventh aspect of the invention there is provided a client workstation in communication with a host server in accordance with the second aspect of the invention.

According to an eighth aspect of the invention there is provided a method of providing knowledge data for purposes such as industrial troubleshooting, wherein elements of knowledge data are owned by different parties, and wherein the data are stored on a plurality of discrete databases such that the data belonging to each party are stored in a separate party- specific database, the method being performed by a server and comprising: establishing a data communications connection, via a network, between the server and a workstation on which the knowledge data is to be viewed; receiving input from the workstation requesting information; establishing a data communications connection, via a network, between the server and the plurality of discrete databases; accessing the plurality of discrete databases for elements of knowledge data in response to the request for information; and displaying said elements of knowledge data from said discrete databases on the workstation display. This advantageously enables different databases to be used to store elements of knowledge data belonging to different owners, thereby preserving intellectual property rights and the ownership of industrial know-how, yet enabling the relevant information to be brought together to present advice to the user in a coherent manner.

Preferably, the workstation displays symptoms associated with an industrial process, and corresponding possible contributory factors, and the method further comprises the user selecting a possible contributory factor and the server accessing the discrete databases for elements of knowledge data which provide information on said contributory factor. For each possible contributory factor for a given symptom, the workstation may further display an indication of the likelihood of the contributory factor being responsible for the given symptom. This guides the user as to which contributory factors should be investigated first, thereby expediting the troubleshooting or help-finding process.

The elements of knowledge data may be displayed on the workstation in response to the user selecting a corresponding hot link on the workstation display. Alternatively the elements of knowledge data may be displayed by populating a predefined table with the elements of knowledge data, said table being displayed on the workstation display.

According to a ninth aspect of the invention there is provided a server configured to execute a method of providing knowledge data for purposes such as industrial troubleshooting in accordance with the eighth aspect of the invention.

According to a tenth aspect of the invention there is provided a computer program executable by processing means to cause a server to execute a method of providing knowledge data for purposes such as industrial troubleshooting in accordance with the eighth aspect of the invention.

According to an eleventh aspect of the invention there is provided a computer program in accordance with the tenth aspect of the invention, stored on a data carrier.

According to a twelfth aspect of the invention there is provided a workstation in communication with a server in accordance with the ninth aspect of the invention.

According to a thirteenth aspect of the invention there is provided a method of retrospectively viewing the behaviour of industrial equipment, the behaviour of the equipment having been recorded by one or more data gatherers logging measured parameters over a period of time and recording these measured parameters in one or more corresponding data items, the method comprising: displaying a pictorial representation of the industrial equipment on a visual display screen, the pictorial representation comprising one or more elements, the position of the elements determined by the values of corresponding measured parameters recorded in corresponding data items; and animating the elements in accordance with changes in the measured parameters with time. This method may be used to play back the behaviour of a single piece of equipment, or a part (or whole) of a process line, including a view of the evolution of the product through the process, and provides an intuitive manner by which to identify faults.

According to a fourteenth aspect of the invention there is provided a computer program executable by processing means to execute a method of retrospectively viewing the behaviour of industrial equipment in accordance with the thirteenth aspect of the invention.

According to a fifteenth aspect of the invention there is provided a computer program in accordance with the fourteenth aspect of the invention, stored on a data carrier.

Brief Description of the Drawings

Embodiments of the invention will now be described, by way of example, and with reference to the drawings in which: Figure I illustrates a network architecture comprising a plurality of industrial data gatherers, a workstation, a server, a plurality of data sources, and a plurality of databases; Figure 2 illustrates a software architecture for the web server and web client, and associated items of network hardware; Figure 3 shows a schematic representation of a process line in a steelworks, together with associated network devices and software architecture; and Figure 4 illustrates the operation of a knowledge base in which elements of knowledge data are obtained from a plurality of discrete databases.

Detailed Description of Preferred Embodiments

The present embodiments represent the best ways known to the applicant of putting the invention into practice. However they are not the only ways in which this can be achieved.

In this description a software toolkit application known as the ALSPA M80i Toolcase will be described, together with a way of using the toolkit so as to provide advanced support for Plant Asset Management. The ALSPA M80i Toolcase is a web-based application for use on networked personal computers (PCs), that allows users to perform searches on one or more data sources. The associated hardware and network infrastructure which support this computer program will also be described.

The ALSPA M80i Toolcase application provides an environment for faulttracing, event analysis and ongoing maintenance, and for quality improvement, which can extend to literally any industrial plant equipment and data-sources that form part of the connected system.

An example of a connected system for implementing the ALSPA M80i Toolcase application is shown in Figure 1. Within an industrial plant there are many data gatherer devices that gather industrial data during the course of the operation of the plant. These are shown schematically as items 110, 112, 114 and 116, and may be standard programmable logic controllers (PLCs), remote input/output devices, standard variable speed drives and other sensors and instruments. These devices 1 10, 1 12, 1 14 and 116 are networked together within the plant by a local network 117, which is typically an Ethernet network, although other types of network are possible.

Also connected to the network 1 17 are control devices 1 18, 120, 122, such as the ALSTOM ALSPA C80-HPCi high performance controller, or other electronic controllers. These in turn are connected to a network 129, which again may be an Ethernet network, or may form part of (or be connected to) a larger network such as a site intranet, a wide area network, or the Internet 131. The data gatherers may be connected directly to the network 129 or the Internet 131. As shown in Figure 1, a notebook computer 125 (or another type of personal computer) may also be used as a data gatherer, and is connected to plant equipment as required to upload process-related data.

Connected to this network 129 are process control computers 124 and 126, programming and engineering control computer 128, and a server 130 which stores such data gathered by data gatherers around the plant as is not stored within the data gatherers themselves. Data gatherers which store data within themselves also function as servers on the network.

Also connected to the wide area network or internet 131 are a number of knowledge-related databases 136, 138, 140 and 142. The purpose and operation of these databases will be described in more detail later.

The ALSPA M80i Toolcase application is hosted on a web server 132, and the user interacts with this using a web client workstation 134, running a web browser such as Microsoft Internet Explorer to provide the user interface. The web server 132 provides a search engine and other functionality that will be described in more detail later.

Figure 2 illustrates in more detail the interrelation between the web client 212 and the web server 210 within the framework of the ALSPA M80i Toolcase application 200. Interaction between the user interface (web client) 212 and the web server and search engine 210 takes place using an HTML script processor 214. The user interface allows the user to define the search criteria and displays the search results. Using the user interface 212, the user can carry out searching and sorting of data items from any of the connected data sources, and can then view or analyse the data items as desired.

Searching and sorting of data items The ALSPA M80i Toolcase application allows users to perform searches on one or more industrial data sources, and view selected data sources. The term "data source" has a broad meaning. In many cases, a data source would be a plant database 240, e.g. containing recent data logged by monitoring devices (e.g. steel thickness data), recent trip histories (i.e. records of faults when plant equipment tripped), reference data of past production results, reference data of past faults, or other types of data which may be of relevance to a particular industry.

The plant database would typically be governed by process and production control systems 242. A data source is not limited to a database, though, and may be a plant data gatherer 238 (e.g. devices 110, 112, 114, 116 and 125 in Figure I - these would also be controlled by control systems 242), and also other sources of data such as database records, individual data files, and so on. Additional data sources can be added to an existing system, to extend the searching capabilities of the ALSPA M80i Toolcase application. The data source types, and the specific data sources, to which a user has access may be limited by a systems administrator.

The user may carry out both time-based searching (e.g. last n days, last n hours, last n events in time) and text-based searching (e.g. against keywords within a data item or header). The user may thus search in terms of the time when data was logged, the location of the plant equipment where the data was recorded, textual content, the identity of the piece of industrial equipment, or the equipment type. For example, the user may request a search of all accessible data sources for data recorded between 10:00 and 11:00 on a given day in which a fault was known to have occurred at some point between those times. If it is known that the fault occurred on a certain piece of equipment, then the search can be limited further to the data sources pertaining to that piece of equipment.

The search engine of the web server 210 processes the search requests and handles the querying of the individual data sources. The search engine may utilise one or more databases 240 in which the configuration details of the data sources, interfaces and consumers are held.

The web server 210 can search across a number of data sources. Each data source requires a component called a Data Source Interface (DSI) 232 to be installed and registered. In addition to this each data source requires a so-called Data Packager to retrieve raw data items from the data source and prepare them for downloading, which is done on demand.

When carrying out a search, the search engine 210 identifies those data items in the connected data sources that satisfy the search criteria, either by finding the required criteria in the identification data of the data items (the identification data being, for example, a job identification code, a time stamp or some other "header" information), or by searching all the data within the data items. The data items that satisfy the search criteria are then listed on the display of the workstation, within the web browser interface 212.

After performing the search, the user has the option to download any number of data items by selecting them within the web browser interface 212. If there is no file to download, then no link will be displayed. Each data item is packaged up with a special MIME-type, the MIME type specifying a data viewer to be used to subsequently view the content of the data item on the workstation. The user may build up a list of data items to be downloaded in a kind of "basket", and the items in the basket can then be downloaded in one operation. The data items may be transmitted in a packaged format, such as in a single ZIP file. This contains all the individual data items, together with their MIME-types. In addition to the packaged data items, an index file (e.g. index.html) may also be included, which lists all the names and identification data (and descriptions if available) of all the data items in the package.

Viewing and analysis of data items The downloaded data items may then be viewed and analysed (e.g. by clicking on a desired data item listed in the search results) to enable the user to identify the cause of a fault, or perform some other kind of analysis. Data viewers 236 form an integral part of the ALSPA M80i Toolcase package, and enable the user to study specific phenomena or ranges of interest (e.g. precise moments in time). Each data viewer requires a so-called Package Handler component 230 to be installed for it to work with ALSPA M80i Toolcase application. The Package Handler component is installed on each client machine that will be downloading files, and is responsible for unpacking the compressed binary file and starting the appropriate viewer with the downloaded file. When a data item is selected, its contents are displayed in an appropriate data viewer 236 (e.g. as specified in the data source's Package Handler) to present the data graphically. Several data items may be viewed simultaneously, using several instances of a given data viewer.

When adding a new data source to the ALSPA M80i Toolcase application, typically a new Data Source Interface 232 and a corresponding Package Handler 230 would need to be installed. Data viewers may also be supplied along with the corresponding data gatherer hardware, or provided by the manufacturers of the data gatherers. Alternatively, third party data viewers may be used to view selected data items. A new data viewer may be provided when a new data source type is added to the ALSPA M80i Toolcase application.

If more than one viewer is supported by a given data source, the user can select which viewer will be used.

A data item may need to be translated from one data format to another format compatible with the selected viewer. The ALSPA M80i Toolcase application is configured to compare the data source with the selected viewer; if translation is required and an appropriate translator exists, it will then process the data by applying the appropriate translator.

Data may also be analysed using one or more data analysers 234, such as Matlab, Famos, Microsoft Excel, etc. These would be installed on the user's workstation as required, along with an associated Package Handler. These enable the user to perform mathematical analyses on the data, such as extrapolating, averaging, carrying out trend analysis or tolerance analysis, etc. Animated data viewer An animated data viewer (referred to as ALSPA M80-A) may also be used to display the data from a given plant or process line. The animated data viewer ALSPA M80-A provides the user with animated pictorial representations of the constituent parts of the process line. For example, for a steelworks the user would see animated rollers, the movement of the steel product, etc., generated using the content of the data items obtained for that process line over a desired time period. The animated data viewer needs to be custom installed for a specific site, to configure the sequence of pictorial representations to correspond with the real process line, and to specify which data items correspond to which pieces of machinery. It may be programmed using Microsoft Visual BASIC. To aid the installation of the animated viewer for a given process line, generic configuration files may be provided that correspond with common pieces of equipment, hot mills, cold mills, process lines etc. These may then be adapted to suit the site in question.

1 5 Templates As has been described, the ALSPA M80i Toolcase application permits the user to search across databases of data items, to sort these data items as desired, and to view selected data items by launchingappropriate viewers. When launching the viewer to view the selected data item, the user may also nominate a Template 224 that will be launched together with the associated viewer. A Template is, in essence, a configuration file which presets the viewing software to a specific configuration (e.g. a particular degree of zoom, pan, selection of certain traces and suppression of other traces within the selected data item, etc.), so as to display the selected data item in a manner that allows the user to gain greatest insight from the data.

Templates may also be used to generate pre-configured reports (e.g. using Microsoft Office components, Matlab, Adobe Acrobat, etc).

Templates are stored within the ALSPA M80i Toolcase connected system, so that any user may access them easily and naturally. The user interface presents the Template function as a natural part of the process of sorting, selecting and viewing data items. The list of Templates available for selection may be modified such that only those Templates appropriate to a specific data gatherer are shown.

If the user adjusts the graphical presentation of a data item (e.g. degree of zoom, captions, etc.) and considers the new layout to be of future use, then he may save this layout as a new Template. This Template can then be applied to other data items in the future, and will present them using the same layout.

Agent Manager An Agent is an automated routine (similar to a macro) that is typically used to collect data from the data sources at given times, or in response to given conditions. The Agent Manager 216 is an extension of the ALSPA M80i Toolcase application which manages the Agents and schedules their jobs. Each job may consist of a single task or a set of linked tasks. Three broad categories of tasks are envisaged that would be performed by Agents and governed by the Agent Manager 216: À To act as a calculator for calculating Calculated Indicators 218, processing data and performing calculations to determine values and/or trends that are not directly measurable. Pre-prepared automated routines 246 (e.g. trend building) may be employed. The results may be stored in databases if required (see below).

À To act as an advisor 220 and to alert people (e.g. by sending out an advisory message 228) when a problem happens, or to launch activities when certain conditions are fulfilled or a critical event is detected. An advisory message 228 could be sent by e mail to a supervisor or process engineer.

To act as a report builder 222. The Agent collects data regularly and generates a report 226, e.g. using a Template 224 and/or a standard application such as Microsoft Word or Excel.

Combinations of these tasks are possible. These results of any of these tasks can be saved to Agent Results databases 244, and are thus added to the range of data available for search and analysis.

The Agent Manager 216 allows manipulation and analysis of a returned record set, by passing the data to an external application (engine), where the execution of a user defined script, if required, will produce results. The results may take any form that the engine can produce i.e. graphs, reports, databases, e-mail alerts etc. The Agents can be set up in a schedule to run periodically, or triggered manually on demand, via the user interface 212 of the ALSPA M80i Toolcase application.

I O Brief Definitions À Agent M anager: A database application that contains the configuration information for Agents and their associated jobs. It is essentially a database of all the configuration information required to run jobs. It may comprise a number of modules that interact to perform the required tasks.

À Agent: An agent controls the data flow of scripts and engines. It defines the input and output for the engines, or may be a stand-alone application.

À Job: A set of parameters that define the search criteria, the script to use and the Agent to launch. A job may comprise a single task or multiple linked tasks.

À Scheduler: Controls the execution of the jobs, allowing the user to specify the frequency and start date/time for individual jobs, e.g. Run every 24 hours 18:00 Run once 19/12/2003 23:45 Run every Friday 23:59 Run when triggered (e.g. by a specified event) Run on completion of another Agent À Engine: This can be a bespoke application or third party product such as Matlab, Microsoft Excel, SQL Server, etc. The engine is used to do the work on a returned recordset to produce the results.

À Script: The script is a set of instructions written in the native language of the associated engine. It is used to control the engine's operation.

The ALSPA M80i Toolcase application acts as the front end for the Agent Manager, allowing the user to input Agents/scripts/jobs, set up the schedules which run the Agents, and to view any results.

Example of the use of the ALSPA M80i Toolcase application with a process line in a steelworks Figure 3 shows a schematic diagram of a process line in a steelworks. The process line comprises reheat furnaces 302, horizontal scale breaker (HSB) 304, a reversing rougher 306, coilbox and crop shear 308, a seven stand finishing mill 310, run out table (ROT) and callers 312.

The performance of the reversing rougher 306 is captured and recorded by data gatherer 316, which logs its data in database 318. The finishing mill 310 is monitored by data gatherer 324 and the data are stored in database 326. Similarly, the callers 312 are monitored by data gatherer 334 and the data are stored in database 336.

Databases 318, 326 and 336 are all connected to the plant Ethernet network 340. Also connected to this network are a database 314 of drive trip data, a database 322 of alarms data connected to the plant's supervisory system 320, a database 330 of HPC (high performance controller) alarm data provided by the HPC system 328, and a database 332 of HPC trip histories.

The ALSPA M80i Toolcase application 342 is connected to the databases 314, 318, 322, 326, 330, 332 and 336 via the network 340. A database 338 of Agent Results is also connected to the network 340, for use by Agents in the ALSPA M80i Toolcase application.

The ALSPA M80i Toolcase application 342 provides the user with the ability to: À search, sort and select (344) data items from the connected databases 314, 318, 322, 326, 330, 332, 336 and 338 À view (346) chosen data items, applying a Template 348 if desired, using graphical viewing software 358, the animated data viewer 360, analysis software 362 or "native" file viewers 364 (e.g. as a web page for alarm data, or a PDF file for reports, etc.) The ALSPA M80i Toolcase application also provides an Agent Manager and scheduler 350, which can be used to: À compute data (352), the results being stored in the database of Agent Results 338 À send advisories (354) À prepare reports (356), optionally applying a Template 348 as appropriate Knowledge Bases The ALSPA M80i Toolcase application, when connected to its respective data sources, includes the components necessary to allow creation of knowledge bases.

A knowledge base is a collection of experience-based information and information-linkages, which allows constructive use to be made of the vast amounts of data which are connected to a system. The knowledge base software is hosted by a web server and accessed by a client A knowledge base typically comprises three aspects: a) The means of linking to and accessing quantities of "raw data" b) A "top-down" means of guiding the user, so that he/she may quickly identify a course of action to respond to a given situation. This is typically driven by Key Performance Indicators (KPI) which relate to the user's needs, and involves a table which shows a relationship between KPIs and their causal factors. This allows the user to select a specific causal factor to be investigated, as his/her course of action in response to a given situation.

c) A means of guiding the user and supporting his chosen course of action, by bringing together as much relevant data as possible, and showing the user where to find the tools and utilities that he/she needs in order to follow the chosen course of action.

Figure 4 shows the broad concept (referred to as ALSPA S80i Toolcase) by which the above three aspects are linked. Here, a KPI dependency table 400 is shown (as would be displayed within the ALSPA S80i Toolcase application on the user's workstation) relating to the width of the steel product in a hot strip mill. Further dependency tables 410, 412, 414 etc. are provided for other KPIs. The columns 420, 422, 424, 426, 428 in the table 400 relate to various problems or symptoms that may be experienced in the hot strip mill concerning the width of the steel product. The rows 430 in the table provide various causal factors that may be responsible for the various problems or symptoms. In the table, the likelihood that a particular factor is responsible for the corresponding symptom is provided as a percentage figure. For example, as stated in cell 432, there is a 30% likelihood that the furnace head-tail temperature is the cause of a problem in the head-to-tail width of the steel product.

If the user wishes to obtain further information relating to the furnace head-tail temperature, then he can click on the cell of interest (e.g. cell 432) in the dependency table. This then brings up a support sheet 440 on the workstation display. (Other support sheets 442, 444, 446 are indicated for other cells in the dependency table.) The support sheet 440 is populated with links to data from a plurality of different databases containing relevant information, e.g. an equipment database 460, production database 462, apples database 464, and further databases such as 466 (also shown as databases 136, 138, 140 and 142 in Figure 1).

The underlying principle at this stage is that the data stored in the databases 460, 462, 464 and 468 (and potentially the databases themselves) are not under common ownership, but instead the information in each database belongs to a different owner, e.g. in accordance with intellectual property rights. For example, the information in the equipment database 460 may be owned by the manufacturer of the plant equipment, whereas the information in the production database 462 may be owned by the owner of the steelworks, and information pertaining to the overall process line and its control systems may be owned by yet another party. By keeping these data in separate databases the ownership of the data is well delineated, the ownership of the intellectual property is preserved, and the owner of each database is able to manage its respective database independently of the others.

The support sheet 440 requested by the user is thereby populated with links to data from the various independent databases. In the support sheet, alongside each of a number of headings (e.g. "Mechanism of effect" 450), a hot link (e.g. 451) is provided to call up the corresponding explanatory information provided by the individual databases. If the corresponding information is brief, then the information may be displayed directly in the support sheet itself, thereby negating the need for the user to follow a hot link.

The knowledge base software for ALSPA S80i Toolcase may be provided separate from the ALSPA M80i Toolcase software which enables searching, sorting and viewing of data items.

There are two complementary possible ways in which knowledge bases can be utilised. The first is for the knowledge base to be built by the client, with assistance from the supplier's experts, such that the knowledge base is specific to the client's plant. The second is for the knowledge base to be built by the supplier, to capitalise on his know-how gained through working on many plants. In this case, the knowledge base may be used to assist many users.

These two possible methods may be combined to suit specific needs.

Summary

The ALSPA M80i Toolcase application is a user-friendly web-browser environment, in which data from many different sources and data-formats can be brought together in a single view.

À Search/sort through ALL data-sources via a single interface À Web-interface for easy LAN (local) or WAN (remote) access À Easy to add extra sources of data, and new data-formats À Auto-calculate derived results, to add value to the data À Identify anomalies/trends which could not previously be seen À Click and launch associated data viewers; export selected data to powerful analysis packages (Matlab, Famos, etc.) À Use this tool to collect/organise know-how, help to buildluse Plant Asset Management (P.A.M.) knowledge databases, as a basis for ALSPA S80i Toolcase Test data or results for the ALSPA M80i Toolcase application are given in Appendix 1. This takes the form of the design test specifications to which software developments of the ALSPA M80i Toolcase application have been tested.

Two examples of case studies to illustrate the use and benefits of the ALSPA M80i Toolcase application are given in Appendix 2.

Appendix 1 - Validation Test Specification

This Appendix describes the validation testing of the ALSPA M80i Toolcase application after installation.

1. I RESOURCES AND TEST ENVIRONMENT These tests are to be performed on a machine using Internet Explorer v5+ or later. A machine running Microsoft Windows 2000 Professional/Server and IIS v5.0, or a machine running Microsoft Windows NT 4 Server/Workstation and IIS v4.0, should host the M80i Toolcase application.

1.2 END-USER TESTS The following tests ensure that the end-user functionality is as expected.

1.2.1 Search Form Functionality

Description

a) Select each of the search types in turn b) Select the two different types of datasource list views c) Select "Specific Date" search, and try each button on calendar d) Enter a valid and invalid date into the input field, and click search e) Select "Time range" search type and try all buttons on calendar f) Select one datasource and perform searches using each of the search types g) Select more than one datasource and perform searches using each of the search types h) Enter different search criteria (both valid and invalid) for each search type and for a datasource of each type.

i) Select "Text" search type and click additional link for making the calendar appear.

j) Enter valid text search strings, this can include * at start or end (or both), or the text search criteria can be entered in double quotes, this should perform an exact match search. Entering a text string alone (without *'s or quotes) will search for the string using wildcards.

k) Enter valid text searches with different time ranges.

Expected Results a) The search panel interface should change according to which search type is selected, the calendar should appear when appropriate and the number of entry

fields should change as appropriate.

b) When the appropriate button is pressed, the datasource list should change.

c) Calendar buttons should all work as expected, i.e. month, year and day buttons should all change.

d) Date should be validated successfully and an error message should appear if appropriate e) Two entry boxes should appear and the calendar buttons should all work 1 5 correctly f) In each case the correct search results should be returned g) In each case the correct search results should be returned, and the search results should be sorted and interleaved as appropriate h) The application should behave sensibly, when search criteria is valid it should return valid search results, and when the search criteria is invalid the application should display an error message.

i) When the "Text" search type is selected, the interface will change to display an additional link. Clicking this link should display the calendar.

j) The search should return the correct results in each case.

k) The search should return the correct results in each case.

1.2.2 Toolbar functionality

Description

a) Move the mouse pointer over the buttons on the toolbar b) Click each button on the toolbar one after the other Expected Results a) As the mouse pointer is moved over each button, the button should change colour and a textual description should appear underneath the button b) The correct page should open after each button is clicked 1.2.3 Downloading items

Description

a) After performing a search, if the search returns more than one item, attempt to download one of the items.

b) Download an item for which there is more than one viewer.

c) Download a file on a machine that does not have the appropriate package handler installed.

d) Download a file and save it to disk (rather than opening it).

e) Repeat test (a) above for different all the different datasource types, in other words exercise each of the package handlers on the client.

Expected Results a) The file should be downloaded and handler by the appropriate package handler and then opened using the appropriate data viewer.

b) The application should display a list of the registered data file viewers and the user must select one of these and then click the "download" button.

c) The standard windows dialog box should appear prompting the user to select a viewer for the file that has been downloaded.

d) The browser will prompt the user for the location of the file, and then it should be saved there. After renaming the file with a.zip extension, the user should be able to open it using an application like Winzip.

e) All files should be successfully downloaded and handled by each of the package handlers.

1.2.4 Downloading multiple items

Description

a) After performing a search, if the search returns more than one item click the check-box alongside the item. Do this for a number of items and then click the "Download Now" link.

b) After performing a search, if the search returns more than one item click the check-box alongside the item. Do this for a number of items and then click the "Copy to My Data" link.

c) After adding some items to the basket, perform another search, if the same items are in the result set, there should be an image of the basket on the left to indicate this. Select some additional items and click the "Download Now" link.

d) Click the "Basket" icon, and remove some items from the basket by selecting them and clicking the "Delete From My Data" link.

e) Click the "Basket" icon, and download some items from the basket page by selecting them and clicking the "Download Selected Items" link.

f) Click the "Basket" icon, and then the "Download All My Data" link.

Expected Results a) The selected items should be retrieved and compressed into a zip file, the file should start downloading after a few seconds.

b) The selected items should be added to the basket, click the basket icon (from the toolbar) to see the contents of the basket, ensure all selected items have been added. As items are added to the basket, the check-box should change to an image of a basket.

c) The selected items should be retrieved and compressed into a zip file, the file should start downloading after a few seconds.

d) Selected items should be removed from the basket and the basket content will be redisplayed.

e) Selected items should be compressed and returned in a zip file. The items should still be left in the basket.

f) All items in the basket should be compressed into a composite zip file and downloaded, the items should be removed from the basket as well. s

1.3 ADMINISTRATOR ROLE TESTS For these tests, the user will need to login as ALSPA M80i Toolcase Administrator, this can be done by selecting "Client Options" from the toolbar and then entering the Administrator password in the appropriate box.

1.3.1 Server options

Description

a) Select default language setting to be a different language and then launch the application on another machine that has not seen the ALSPA M80i Toolcase 1 5 before.

b) Change the Administrator password using the "Change Admin Password" link.

c) Change the default date format for the server and then launch the application on another machine that has not seen the ALSPA M80i Toolcase before.

Expected Results a) The language selected should be the default for all new clients that use the application.

b) The new password should be successfully saved.

c) The date format should be the default on all new clients that use the application.

1.3.2 Datasource Types Assuming you have logged in as an Administrator.

Description

a) Click on each of the datasource types to display all the parameters b) Click each link on the page, to ensure all links work and do not return errors c) Attempt to edit one of the datasource types d) Add a new datasource type e) Delete a datasource type f) Add a new parameter to an existing type g) Edit, delete an existing parameter h) Add a new translation script and then delete it i) Register a new data viewer, then set up another datasource type to use the new data viewer (using "Select existing viewer" option) j) Delete the data viewer Expected results All the above tests should work as expected, and should not give any error messages.

1.3.3 Datasource Instances

Description

a) Create a new instance of one of the datasource types, ensure that the parameters listed are correct for the datasource type selected b) Attempt to add a new instance, without filling in all the required fields. Add should fail c) Edit parameters of existing instances

d) Edit an existing instance name and description

e) Delete an existing instance Expected Results All the above tests should return correct results, or should return appropriate error messages.

1.4 AGENT MANAGER TESTS Ensures that the Agent Manager integrates correctly with the ALSPA M80i Toolcase

Description

a) Ensure that new Agents and new scripts can be added, and that new jobs can be configured.

b) Ensure that ALSPA M80i Toolcase can be used to set up various schedules for the Agent Manager.

c) Ensure that ALSPA M80i Toolcase can display a list of the scheduled jobs.

d) Ensure that the scheduler triggers jobs correctly.

e) Ensure that a job can be triggered manually from ALSPA M80i Toolcase.

f) Ensure that the ALSPA M80i Toolcase can display the result information from the Agent Manager.

Expected results a) The new Agent should be added to the Agents list, the new script should be added to the scripts list, and the new job should be added to the jobs list.

b) & c) The new schedules should be listed in the scheduled jobs list.

d) Job should run when scheduled e) Job should run when triggered f) The Result information should be displayed for the executed jobs.

Appendix 2 - Case Studies The following examples are intended to convey an impression of the power and versatility of the ALSPA M80i Toolcase application and its companion tools. These scenarios show how a user can set up and use the ALSPA M80i Toolcase application, and cover a range of applications in the metals, marine and drives industries. This is just a small selection of the situations and industries in which the ALSPA M80i Toolcase application can be used as the basis for problem solving.

Case Study 1: Hot Strin Mill Nature of Problem A regular problem was caused by unexplained Finishing Mill main-drive trips, about once per day. Each time this happened, there were 3 significant losses: 15. Product upstream in the mill was lost Production time was lost while drive was re-started À Production rate was slowed down afterwards, since no-one knew what was causing the trips Relevant Data-sources À Drive trip data (logged by ALSTOM controllers and archived to disk).

À Logged data from the process control system (set-up data for guides, screw gaps, speeds, looper angles, etc: and dynamic strip data during rolling, from gauges, pyrometers, loadcells etc.; all captured by ALSPA M80-M monitoring software and archived to disk).

Alarms data from the HMI system (archived to disk).

Analysis Procedure Use ALSPA M80i Toolcase search/sort functions to search through lists of drive trips and coil-traces captured when rolling a range of different products.

À Look for coincidences: identify all drive trips while threading or rolling a strip, and match each drive-trip file with the associated striptrace.

À Examine each pair of files, using file-viewers launched automatically from ALSPA M80i Toolcase.

À Using both traces, in each case try to identify whether some problem in the strip caused the drive trip, or whether the drive trip happened first and caused a strip problem. (NB: ALSPA M80-A animated data viewer was also used to view strip traces in animation mode, to identify exactly what was happening in the mill when the drive tripped.) Analysis Results The investigation showed that the main drive of Stand 6 was tripping about once per day, always due to over-voltage after a sudden loss of stand load. It was seen that whenever this happened, the product being rolled was a thin-gauge strip of hard material, of a type introduced since the control system was originally set up. With this data as guidance, specific changes were made to the drive set-up, and the problem was completely eliminated.

Benefit to Client The problem had occurred roughly once per day, each time involving the loss of at least one bar upstream plus a lO-minute (minimum) delay to re-start the drive, plus a further production delay due to slower production rate for a few hours after the fault. The saving on this mill, which rolls 21 shifts/week, was: Avoid losing about 300 bars per year scrapped from upstream queue equating to about $800k/year (based on $90/ton and 30 ton/coil).

À Avoid losing 50 hours/year due to drive re-starts, and a further 30-40 hours/year due to production rate slowdown -1-2% of total production time.

Case Study 2: Marine Drives Nature of Problem À Any marine vessel using electric drives for propulsion and/or processing (e.g. drilling, pumping, positioning) must also generate its own electric power. The power system will be less "stiff" and potentially more unstable than for a land-based installation.

À Power system instability may cause large drives to trip e.g. due to transient over voltage. On the other hand, sudden switching of a large drive may also cause instability in the power system, due to the relatively slow response of large diesel generator sets.

À In a typical scenario, a large drive may trip, causing one or more generators to trip due to reverse power flow; any remaining generators are now overloaded, and trip in turn, causing a complete power blackout.

À What was the root cause? Did the drive trip initiate the power system instability; or l 5 was it instability in the power system which first caused the drive to trip? Relevant Data-sources À Drive trip data (logged and archived to disk) À Logged data from the power generation system (captured by ALSPA M80-M and archived to disk) À Alarms data from the HMI (human/machine interface) system, archived to disk Analysis Procedure À Search databases of drive system and power system trips, and HMI alarm logs.

25. Use ALSPA M80i Toolcase to look for coincidences between trips in drive system and power system; examine each pair of trips; launch new searches, to identify relevant data from other systems (e.g. pumps) that may help to diagnose those specific trips.

Analysis Results À It might be noticed (for example) that a large pump created a switching surge on the power system, which caused a main propulsion drive to trip at a time when power system loading was heavy.

À To anticipate future similar events, use the "Calculated Indices" function of ALSPA M80i Toolcase to initiate specific monitoring/trending for that combination of circumstances. Avoid a future repeat situation.

Benefit to Client À A fleet operator may typically see 1 power blackout per vessel per year plus many power "brown-outs". À A blackout will bring production to a halt for at least 1 hour, at a

minimum cost of $10k-$20k in charter fees, manpower, fuel etc., and losses may run into millions if critical operations are endangered À A brown-out has less effect, but as there are more of them the cumulative cost can be just as high.

À Halving the incidence of such problems, by improved diagnosis and prevention, could save a 10-vessel fleet operator typically $100k-$200k every year.

Claims (43)

1. A method of searching and viewing industrial data items stored in one or more remotely-distributed industrial data sources, a data item comprising: industrial process-related data, identification data by which the data item can be identified, and searchable data representative of industrial process criteria; the method being implemented by a host server and comprising: establishing a data communications connection, via a network, between the server and a client workstation on which the data items are to be viewed, and between the server and said one or more remotely-distributed data sources; initiating a search of the data sources by processing data specifying industrial process or other criteria by which the data sources are to be searched; interrogating the said one or more data sources to identify data items which satisfy the specified search criteria; transmitting to the client workstation data representative of those data items which satisfy the specified search criteria; and enabling a user to view the content of one or more of those data items which satisfy the specified search criteria.

2. A method as claimed in Claim 1, wherein the step of initiating a search of the data sources by processing data specifying industrial process or other criteria comprises receiving input from the user, via the client workstation and the network, specifying the desired criteria.

3. A method as claimed in Claim 1, wherein the step of initiating a search of the data sources by processing data specifying industrial process or other criteria comprises running an automated agent, which specifies pre-determined criteria and initiates pre determined actions.

4. A method as claimed in any of Claims 1 to 3, wherein the step of enabling a user to view the content of those data items which satisfy the specified criteria comprises: displaying on the workstation the identification data of those data items which satisfy the specified criteria; S enabling the user to select a data item of interest; and displaying the content of the selected data item on the workstation.

5. A method as claimed in Claim 4, further comprising enabling the user to select a plurality of data items of interest, and downloading those data items to the workstation in a packaged form such as a ZIP file.

6. A method as claimed in Claim 4 or Claim S. wherein the step of displaying the content of a selected data item further comprises using a pre-prepared template, the content of the selected data item being displayed in accordance with conditions defined in the template.

7. A method as claimed in Claims 4, 5 or 6, wherein the step of displaying the content of the selected data item comprises displaying the content graphically.

8. A method as claimed in Claim 7, further comprising displaying the content of the selected data item in an animated form to show changes in measured data over time.

9. A method as claimed in any preceding claim, further comprising receiving input from the client workstation specifying data source types to be searched.

10. A method as claimed in any preceding claim, further comprising receiving and installing a data source interface or driver, to enable a hitherto unrecognised type of data source to be recognised and searched.

11. A method as claimed in any preceding claim, further comprising installing software to enable a hitherto unrecognized type of data item to be recognised and viewed.

12. A method as claimed in any preceding claim, further comprising receiving input from the workstation specifying data sources or data items to be searched.

13. A method as claimed in any preceding claim, further comprising receiving input from the client workstation specifying one or more variables by which to search, the variables being selected from a group comprising: time, location, text content, industrial equipment identity, product-type.

14. A server configured to execute a method of searching and viewing industrial data items stored in one or more remotely-distributed industrial data sources, a data item comprlsmg: industrial processrelated data, identification data by which the data item can be identified, and searchable data representative of industrial process criteria; the method comprising: establishing a data communications connection, via a network, between the server and a client workstation on which the data items are to be viewed, and between the server and said one or more remotely-distributed data sources; initiating a search of the data sources by processing data specifying industrial process or other criteria by which the data sources are to be searched; interrogating the said one or more data sources to identify data items which satisfy the specified search criteria; transmitting to the client workstation data representative of those data items which satisfy the specified search criteria; and enabling a user to view the content of one or more of those data items which satisfy the specified search criteria.

15. A server as claimed in Claim 14, further configured to execute a method as claimed in any of Claims 2 to 13.

16. A data communications network comprising one or more remotelydistributed industrial data sources and a server as claimed in Claim 14 or Claim 15.

17. An industrial plant or process line comprising a plurality of data gatherers configured to gather industrial process-related data, said data being stored in one or more remotely-distributed industrial data sources in communication with a data lO communications network as claimed in Claim 16.

18. A computer program executable by processing means to cause a server to execute a method of searching and viewing industrial data items stored in one or more remotely-distributed industrial data sources, a data item comprising: industrial process-related data, identification data by which the data item can be identified, and searchable data representative of industrial process criteria; the method comprising: establishing a data communications connection, via a network, between the server and a client workstation on which the data items are to be viewed, and between the server and said one or more remotely-distributed data sources; initiating a search of the data sources by processing data specifying industrial process or other criteria by which the data sources are to be searched; interrogating the said one or more data sources to identify data items which satisfy the specified search criteria; transmitting to the client workstation data representative of those data items which satisfy the specified search criteria; and enabling a user to view the content of one or more of those data items which satisfy the specified search criteria.

19. A computer program as claimed in Claim 18, wherein the method further comprises a method as claimed in any of Claims 2 to 13.

20. A computer program stored on a data carrier, said computer program being executable by processing means to cause a server to execute a method of searching and viewing industrial data items stored in one or more remotely-distributed industrial data sources, a data item comprising: industrial process-related data, identification data by which the data item can be identified, and searchable data representative of industrial process criteria; the method comprising: establishing a data communications connection, via a network, between the server and a client workstation on which the data items are to be viewed, and between the server and said one or more remotely-distributed data sources; initiating a search of the data sources by processing data specifying industrial process or other criteria by which the data sources are to be searched; interrogating the said one or more data sources to identify data items which satisfy the specified search criteria; transmitting to the client workstation data representative of those data items which satisfy the specified search criteria; and enabling a user to view the content of one or more of those data items which satisfy the specified search criteria.

21. A computer program stored on a data carrier as claimed in Claim 22, wherein the method further comprises a method as claimed in any of Claims 2 to 13.

22. A client workstation in communication with a host server as claimed in Claim 14 or Claim 15.

23. A method of providing knowledge data for purposes such as industrial troubleshooting, wherein elements of knowledge data are owned by different parties, and wherein the data are stored on a plurality of discrete databases such that the data belonging to each party are stored in a separate party-specific database, the method being performed by a server and comprising: establishing a data communications connection, via a network, between the server and a workstation on which the knowledge data is to be viewed; receiving input from the workstation requesting information; establishing a data communications connection, via a network, between the server and the plurality of discrete databases; accessing the plurality of discrete databases for elements of knowledge data in response to the request for information; and displaying said elements of knowledge data from said discrete databases on the workstation display.

24. A method as claimed in Claim 23, wherein the workstation displays symptoms associated with an industrial process, and corresponding possible contributory factors, the method further comprising the user selecting a possible contributory factor and the server accessing the discrete databases for elements of knowledge data which provide information on said contributory factor.

25. A method as claimed in Claim 24 wherein, for each possible contributory factor for a given symptom, the workstation further displays an indication of the likelihood of the contributory factor being responsible for the given symptom.

26. A method as claimed in any of Claims 23, 24 or 25, wherein the elements of knowledge data are displayed on the workstation in response to the user selecting a corresponding hot link on the workstation display.

27. A method as claimed in any of Claims 23, 24 or 25, wherein the elements of knowledge data are displayed on the workstation display by populating a predefined table with the elements of knowledge data, said table being displayed on the

28. A server configured to execute a method of providing knowledge data for purposes such as industrial troubleshooting, wherein elements of knowledge data are owned by different parties, and wherein the data are stored on a plurality of discrete databases such that the data belonging to each party are stored in a separate party-specific database, the method comprising: establishing a data communications connection, via a network, between the server and a workstation on which the knowledge data is to be viewed; receiving input from the workstation requesting information; establishing a data communications connection, via a network, between the server and the plurality of discrete databases; accessing the plurality of discrete databases for elements of knowledge data in response to the request for information; and displaying said elements of knowledge data from said discrete databases on the workstation display.

29. A server as claimed in Claim 28, further configured to execute a method as claimed in any of Claims 24 to 27.

30. A computer program executable by processing means to cause a server to execute a method of providing knowledge data for purposes such as industrial troubleshooting, wherein elements of knowledge data are owned by different parties, and wherein the data are stored on a plurality of discrete databases such that the data belonging to each party are stored in a separate party-specific database, the method comprising: establishing a data communications connection, via a network, between the server and a workstation on which the knowledge data is to be viewed; receiving input from the workstation requesting information; establishing a data communications connection, via a network, between the server and the plurality of discrete databases; accessing the plurality of discrete databases for elements of knowledge data in response to the request for information; and displaying said elements of knowledge data from said discrete databases on the workstation display.

31. A computer program as claimed in Claim 30, wherein the method further comprises a method as claimed in any of Claims 24 to 27.

32. A computer program stored on a data carrier, said computer program being executable by processing means to cause a server to execute a method of providing knowledge data for purposes such as industrial troubleshooting, wherein elements of knowledge data are owned by different parties, and wherein the data are stored on a plurality of discrete databases such that the data belonging to each party are stored in a separate party-specific database, the method comprising: establishing a data communications connection, via a network, between the server and a workstation on which the knowledge data is to be viewed; receiving input from the workstation requesting information; establishing a data communications connection, via a network, between the server and the plurality of discrete databases; accessing the plurality of discrete databases for elements of knowledge data in response to the request for information; and displaying said elements of knowledge data from said discrete databases on the workstation display.

33. A computer program stored on a data carrier as claimed in Claim 32, wherein the method further comprises a method as claimed in any of Claims 24 to 27.

34. A workstation in communication with a server as claimed in Claim 28 or Claim 29.

35. A method of retrospectively viewing the behaviour of industrial equipment, the behaviour of the equipment having been recorded by one or more data gatherers logging measured parameters over a period of time and recording these measured parameters in one or more corresponding data items, the method comprising: displaying a pictorial representation of the industrial equipment on a visual display screen, the pictorial representation comprising one or more elements, the position of the elements determined by the values of corresponding measured parameters recorded in corresponding data items; and animating the elements in accordance with changes in the measured parameters with time.

36. A computer program executable by processing means to execute a method of retrospectively viewing the behaviour of industrial equipment, the behaviour of the equipment having been recorded by one or more data gatherers logging measured parameters over a period of time and recording these measured parameters in one or more corresponding data items, the method comprising: displaying a pictorial representation of the industrial equipment on a visual display screen, the pictorial representation comprising one or more elements, the position of the elements determined by the values of corresponding measured parameters recorded in corresponding data items; and animating the elements in accordance with changes in the measured parameters with time.

37. A computer program stored on a data carrier, said computer program being executable by processing means to execute a method of retrospectively viewing the behaviour of industrial equipment, the behaviour of the equipment having been recorded by one or more data gatherers logging measured parameters over a period of time and recording these measured parameters in one or more corresponding data items, the method comprising: displaying a pictorial representation of the industrial equipment on a visual display screen, the pictorial representation comprising one or more elements, the position of the elements determined by the values of corresponding measured parameters recorded in corresponding data items; and animating the elements in accordance with changes in the measured parameters with time.

38. A method of searching and viewing industrial data items substantially as herein described with reference to and as illustrated in any combination of the accompanying drawings.

39. A method of providing knowledge data for purposes such as industrial troubleshooting substantially as herein described with reference to and as illustrated in any combination of the accompanying drawings.

40. A method of retrospectively viewing the behaviour of industrial equipment substantially as herein described with reference to and as illustrated in any combination of the accompanying drawings.

41. A server substantially as herein described with reference to and as illustrated in any combination of the accompanying drawings.

42. A workstation substantially as herein described with reference to and as illustrated in any combination of the accompanying drawings.

43. A computer program substantially as herein described with reference to and as illustrated in any combination of the accompanying drawings.