Find your documents from equipment tags.
In most cases, when a service technician or operator searches a document is that he faces a process or equipment malfunction.
The technician must then find the documentation of the equipment concerned as quickly as possible in order to limit the effects of a prolonged outage.
It is for this reason that, on production plants, research master search key is the code equipment (tag).
To find a document relating to one or more equipment, we must have realized this association tag-> document (s) for each of the equipment and for each of the documents. This work is considerable and it is rarely done manually.
Some EDMS editors explain that it is possible to retrieve a document from the moment it contains tags that will be found by the search engine. However equipment tags can be written in different ways such as for example: 00LAC21AP001, 00 LAC21 AP001, OOLAC 21AP001,… For these reasons, the full-text search engines are relatively inefficient for searching of tags in documents.
This is why we have developed an algorithm for the extraction of tags ‘ intelligent ‘who works from the understanding of the structure of the tag and not by similarity.
Autolink allows to extract all tags of a document and provide a consistent index from a base of reference tags. It also helps users identify the tags that are present in the documents but by in the reference database and thus enrich the latter.
Autolink is an optional module of the E – DMS pack which is materialized by an extra tab at the document level and an analysis module. The module operation procedure is as follows:
Step 1: Selecting applicable coding systems for the recognition of the tags in the text of the documents
Step 2: Analysis of the document with Autolink
Step 3: Processing of the results
Case N ° 1: the confidence index is up (green), then the tag contained in document matches exactly (separators pres) tag in the reference database (functional tree).
Case N ° 2: the confidence index is medium (orange), then the tag contained in document looks like the tag in the reference database (functional tree). It is often in this case of a typing error in the document.
Case N ° 3: the confidence index is low (red), then the tag contained in document does not resemble any tag in the reference database (functional tree). It is often in this case a tag which has not been incorporated in the basis of reference or a tag removed/modified during the design process.
It obviously required that documents have a text layer and tags in their content in order to have the greatest efficiency.
As a general rule we have an average of 20% of the installation documents that contain tags, which will cover 80% of the equipment. In other words, there is at least a document associated with a tag for 80% of the equipment after Autolink processing.