Groves For Applications   Table of contents   Indexes   XML-Enabling Enterprise Databases to Simplify Internet Applications
 

A pragmatic approach to IETMs for medium sized projects within the NATO Maintenance and Supply Agency (NAMSA)
 Alan   Moss
  NAMSA  2 rue de la gare
Capellen   L-8302  Luxembourg
Phone: +352.3063.6033
Fax: +352.30.90.09
Email: le-td@namsa.nato.int
 
Biographical notice:
 
AM started his career with 10 years in the UK Royal Navy as an aircraft technician, leaving with the rank of Chief Petty Officer. He continued with aircraft work for some years, both with the Royal Saudi Arabian Air Force as a technician instructor, and subsequently with the British Aircraft Corporation.
 
From '74 to '83 he was employed by the Institute Laue-Langevin, in Grenoble, as a member of the nuclear plant operation and maintenance team.
Bertinchamps, Jean-Marc
Electronic Data Processing s.a.
 Luxembourg  
Olm
 
For the last fifteen years AM has been a civilian staff member of the NATO Maintenance and Supply Agency (NAMSA), in Luxembourg. As part of the NAMSA Air Defence Programme, he has spent many years authoring and editing technical manuals, and controlling contract deliverables. In more recent years he has been heavily involved in acquisition logistics, specifying requirements for technical documentation, and applying Integrated Logistic Support and CALS principles.
 Jean-Marc   Bertinchamps
  Electronic Data Processing s.a.  41 Avenue Grand-Duc Jean
Olm   L-8323  Luxembourg
Phone: +352.305.657-1
Fax: +352.305.657-500
Email: jmbertinchamps@edpsa.com
 
Biographical notice:
 
JMB obtained his degree as an Electromechanical Engineer from the "Université de Liège" in '83. Two years later he obtained a further degree in Software Engineering, also from the "Université de Liège". In '91 he got his MBA from "European University Brussels".
 
After having started his career as a software consultant for different companies such as SIEMENS and CARREFOUR, JMB was appointed as technical support manager for Interleaf Benelux. He left the company to found C.I.S., a Belgian company, two years later he started E.D.P. S.A. a Luxemburgish company active in the publishing and automation of the publishing area.
 
Today E.D.P. is a supplier for major international organizations such as the European Commission, the Court of Justice of the European Commission and NAMSA as well as different major private banking companies.
 
ABSTRACT:
 
Radars are complex and expensive items of equipment. The availability and performance of these systems throughout their life cycle (25-30 years) depends heavily upon how well they are operated and maintained. Effective operation and maintenance in turn relies upon valid, rapidly accessible data in the form of operating procedures, preventive and corrective maintenance procedures, and reference information. Such data must be structured to reflect equipment breakdown and information content, and must identify specific configuration states.
 
One task of the NAMSA Air Defence Programme is to acquire/generate, maintain and distribute documentation that meets the above needs.
 
To help confront the situation of multiple equipment types, multi-national low quantity deliveries, and diverse suppliers, efforts have been made over the years to standardize documentation requirements.
 
Additional factors such as frequent updates, variable configurations, and site-specific aspects have motivated the move towards electronic data techniques. The ability to minimize data redundancy, generate data collections for specific end users, and distribute data electronically in Interactive Electronic Technical Manual (IETM) form underlies our aims.
 
To date, minimum requirements based upon ISO standards have been established, essentially SGML & CGM. A "slim SGML" approach was adopted to ensure a tight structure control but at the same time enable cost effective conversion of legacy data. By its nature the DTD, with only minor adjustment, is suitable for XML application.
 

INTRODUCTION
 
The mission of the NATO Maintenance and Supply Agency (NAMSA) is to provide acquisition support, maintenance management, engineering support, and procurement services to NATO member nations, on an individual and group basis. The Air Defence Programme, within NAMSA, performs this mission specifically for Air Defence Ground (AD(G)) equipment.
 
Part of the Air Defence Programme's task is to provide documentation support for the designated environment, consisting of over 90 operational sites located in 12 nations. The equipment on these sites currently consists of 13 different radar systems and over 40 different display, data handling, and communications systems. No two sites are identical. In documentation terms this equates to the control and maintenance of approximately 2000 master technical manuals, originating from more than a dozen contractors in various formats, and with distribution quantities varying between 50 and 150 copies.
 
The situation is by no means static, some 20 radars of various types are either currently under procurement, or acquisition planning, for future installation within the supported environment. Our task is to ensure that the technical documentation procured with these systems, is acceptable to the nations, meets NATO requirements, and can be efficiently maintained throughout the service life of the equipment.
 
A typical technical manual set for a radar system is shown in Figure 1.


Figure 1. Typical Technical Manual set for a Radar System

  
 

AIR DEFENCE ACQUISITION ENVIRONMENT
 
From the introduction above, it is evident that the acquisition environment does not readily lend itself to a standard approach for optimizing centralized documentation support. The high number of low quantity contracts, multiple acquisition authorities, diverse system maintenance concepts, and the application of international competitive bidding, have all to be taken into account when attempting to standardize.
 
Over the years NAMSA has provided guidance to the acquisition authorities by way of a "requirements reference document" (TME-2506). This document has recently been radically amended to implement NATO policy regarding the acquisition of Interactive Electronic Technical Manuals (IETMs).
 

IETM DEFINITION
 
"Classes" of "IETMs" are often referred to; in fact these are classes of automated TM presentation systems, one of which, class 4, is the generally recognized IETM definition. These are however, at least in their origin, US definitions, and the class 4 definition normally stipulates the use of US-MIL Specs. Furthermore, as often stated in CALS papers, these classes are not hard and fast, or mutually exclusive and the term by itself should not be used for contracting purposes. A more generic definition is needed if the classes are to be referred to independent of US / AECMA or other specifications.
 
TME-2506 is appropriate for the acquisition of "IETMs" in so far as it requires all technical manual source data to reside in a common data repository (SGML or valid XML structured), which has a defined non-redundancy level, interrelated text and graphic elements with attributes, and internally stored interactive features. Distribution IETMs are extracted from this database to meet the needs of individual Nations/sites/end users. The user interface requirements are adequate to ensure that the end users (operation and maintenance personnel) may perform all activities using the electronic media only.
 
The above requirements equate generically to the class 4 definition without imposing, or excluding, the use of US-MIL or AECMA specifications, which may be proposed by the bidder. Eventual use of these specifications will primarily affect the IETM database maintainer, as opposed to the IETM end user.
 
The requirements may also be used for the acquisition of hard-copy technical manuals (with source files in SGML or valid XML & CGM). This will readily permit lower classes of automated TM presentation (ETMs) to be generated (to supplement the continued use of hard copies). Transition to IETMs would of course also be facilitated.
 

TOOLS AND STANDARDS
 
High performance off-the-shelf document management systems provide sophisticated means for controlling workflow, shared access etc. suitable for a large publishing house, but are excessive for our needs. On the other hand, such systems would still require considerable development and customization (assuming this to be possible) to meet our specific environment, and would be far from an "out of the box" solution.
 
If "the system" is set aside for the moment, and one considers standards application, the question of tailoring arises. Even if established US-MIL or European (AECMA) standards and specifications for IETMs are invoked, they can not exactly or completely reflect our needs without further specification. For example the need to define a standard numbering system and information codes, or the need to develop additional content specific DTDs. Furthermore, these "heavy" standards are designed primarily for large projects, involving new-start systems; and are therefore inappropriate for projects involving three or four existing or derivative design radars, with a host of legacy documentation in a variety of formats.
 
Whatever the solution envisaged, it inevitably leads to an extensive development of new contractual requirements, which must be sufficiently concise for generating bid cost data, and for product test and acceptance purposes. Keeping ahead with technology and standards can bring real benefits, but not merely by application, rather as a result of the rethink and redesign of information handling, with cost, and resources constantly in mind in order to improve data validity, access and exchange.
 

ACQUISITION APPROACH
 
From what has been discussed in the previous paragraph, it is clear that neither the specifying of proprietary documentation management systems nor the invoking of available standards constitute attractive solutions. Even if the necessary customization and tailoring needed is concisely reflected in an Invitation for Bid (IFB), there is every likelihood that the contract cost will increase due to mandatory redesign of existing methods, investment in new tools and training, and legacy data conversion issues that need to be addressed by the contractor.
 
Our approach is therefore to contractually define what has to be accomplished, not how. This is achieved by way of generic "performance requirements", which ensure that the deliverable end products can meet test and acceptance criteria in all cases (i.e. whatever the production methods). The criteria must enable purchaser control of the products, ensure similarity of look and feel to end users, and permit life cycle maintenance of the documentation.
 
Performance requirements for use in an IFB include the following topics:
  •  Security (data repository and distribution IETM)
  •  Standards (SGML or valid XML & CGM)
  •  Data Structure
  •  Production System - including a data repository
  •  Links
  •  Distribution IETM
  •  User Environment
  •  User Interface
  •  Prototyping
  •  Test
 
In a further drive to reduce up-front cost and simultaneously reduce the subsequent in-house follow-on maintenance costs, the decision was taken to implement an in-house IETM production system, which will be made available to other AD(G) support facilities, Nations, and contractors. Information on how this system is being developed constitutes the technical portion of the presentation given in conjunction with this paper.
 
Under the performance requirements in the IFB the topics covering "data structure" and the "production system", include options for the use of a purchaser furnished DTD and a purchaser furnished production system respectively; details of which are annexed to the IFB.
 

OVERVIEW OF USER REQUIREMENTS AND STRUCTURE
 

User Requirements
 
Although effort is often concentrated to a large degree upon underlying IETM production methodologies, and control and maintenance of data, success ultimately depends upon end user acceptance of the product. End user access modes, views and functions rely heavily upon the underlying data structure; therefore the basic end user requirements must be clearly defined up front. Access to a wide range of connected information levels and types via a single user interface, and flexibility in working with procedural data, are considered essential for the AD(G) environment in order to match wide ranging technician training and responsibilities.
 
Priority is given to the allocation of system resources to rapid information access needed by organizational level personnel to ensure minimum downtime, and optimum operational state of the system.
 
A common set of user interface components and presentation conventions is used to provide a consistent user interface across all presentation devices. Mandatory user management, control, access and interaction functions are defined, independent of the IETM data repository structure employed, and the precise methods of screen presentation.
 
Since in-depth coverage of this subject is not possible within the limits of this paper, an attempt has been made to graphically summarize our approach to end user requirements, as shown in Figure 2.


Figure 2. User Requirements Overview

  
 

Structure
 
Figure 3 shows the higher level XML elements and some main attributes, and relates these to source and destination environments.
 
The source environment consists of the operational sites, and the installed equipment. The hierarchical structure (with minimum attributes) permits the "primitive" elements, i.e. a paragraph (any level), a table, a procedure, or a graphic, to be identified as "belonging to" a specific: equipment, site or sites, information type and sub-type, topic and sub-topic.
 
The destination environment consists primarily of system managers, technicians, and operators at the radar sites. Hierarchically structured information "sets" can be extracted for specific end users under the "collection" element. The content of the set will determine the user main menu construction, which combines information levels, primary equipment groups, and entry modes. The direct access/search functions may be invoked at various levels in order to optimize information feedback in function of the end user's knowledge of the equipment and specific needs. The actual "names" of information types/sub-types and topics/sub-topics are defined external to the DTD and entered as attributes, this ensures that the structure may be used for multiple technical manual types.


Figure 3. Structure Overview

  
 

Additional Integrated Functions
 
Many of the advantages of IETMs, both for data maintenance and for the user, are only realized if broader integration (or interfacing) is implemented. This imposes the crossing of boundaries inherent to technical manuals, both with respect to the scope of accessible information, and the end user functionality. Essentially this consists in providing logistic functions other than traditional access to data for performing operations and maintenance actions.
 
At present for IETMs, we are concentrating upon a database (or database tables) approach for Illustrated Parts Breakdown lists, and a similar approach to support maintenance management (to handle preventive maintenance planning, recording, and reporting). In addition, direct access to corrective data, via message code entry into a dialogue box is provided for. It is evident that some users will consider it logical to extend such functions to include provisioning activities, recording and reporting of corrective maintenance for fault analysis, etc.
 
The delivery of Logistic Support Analysis Record (LSAR) databases is now the normal practice for AD(G) systems, and as such progressively more IETM data, in particular table data will be derived from these databases. The integration of sub-sets of LSAR relational data with the IETM object data will ensure direct access by the IETM user to valid equipment configuration and logistics information. Some 40 relational data elements are identified to support parts identification and location, maintenance actions, and tools, test equipment, and materials identification and use. These elements are under project configuration control, and the majority are also used for logistics products other than IETMs, thus ensuring overall data integrity.
 

APPLICATION TO NATO PROJECTS
 
The evolution of NAMSA's approach to IETMs and the application to NATO projects are shown in Figure 4.
 
In order to acquire IETMs for legacy data systems, an SGML DTD was developed some three years ago, and is being used by a contractor to produce IETMs for one NATO Nation (3 radars), with possible expansion to three Nations (7 radars). Concurrent with this work NAMSA is also involved with the application of tailored US-MIL Specs under a separate project. The combined experience has been used to update the NAMSA "requirements document" (TME-2506), which was issued for general use by NATO Nations and Agencies in May 1998.
 
The arrival of XML coincided with the start of NAMSA's in-house conversion and production system activities, details of which are provided in subsequent paragraphs. In parallel with this work, TME-2506 is being further updated to include details of the valid XML structure and the production system, and LSAR requirements. In addition the approach is currently being applied to a National project to convert legacy data to XML/CGM for on-line access.
 
The updated requirements will be used as the basis for the acquisition of documentation for several new radar projects in the near future.


Figure 4. Application to NATO projects

  
 

IETM PRODUCTION SYSTEM DEVELOPMENT
 
The following activities were carried out in order to establish the methods and tools to be used.
 

Determine documentation environment
 
Three main steps were performed:
  •  Analysis of existing documentation structure and content,
  •  Investigation into end user requirements,
  •  Review of existing standards and specifications.
 

Determine basic criteria for data structure
 
The underlying information structure and attributes (database) must:
  •  Permit any AD(G) information types/sub-type (manual) to be generated,
  •  Facilitate the importation of legacy data,
  •  Permit specifically identified data to be extracted for update or distribution purposes, e.g. by equipment type, information type, site location,
  •  Identify objects to the degree necessary for applying behavioral rules for display and functions,
  •  Form the basis for user menus permitting access via system breakdown, information structure, and specific modes such as 'safety summary', 'alarm/fault/status' input.
 

Determine approach
 
The basic concept is for a system that:
  •  Is not tied to any one product, operating system, or computer,
  •  Uses COTS products (or even better, those freely available on Internet),
  •  Permits products to be interchanged (e.g. databases; by using standard language for data access)
  •  Allows users to choose their own editor tools (i.e. editor totally independent of the authoring `engine'),
  •  Provides control, validation, and security processes,
  •  Is organized for implementation of a workflow system.
 

Implementation strategy
 
For better ongoing proof of concept it was decided to link the development of the system to an actual project for the conversion of legacy data (a mix of diverse electronic formats and paper) to IETM. The project entails the conversion of some 50 technical manuals for a single radar system installed at 10 sites in 4 nations.
 

IETM PRODUCTION SYSTEM
 
The following paragraphs provide a lead-in to the topics that will be developed during the presentation. Figure 5 illustrates the overall conversion and construction process.


Figure 5. Conversion and Construction process

  
 

Legacy data
 
A major work area has been the integration of legacy data, part of it being available only on paper. Different techniques have been used such as scanning, scanning and OCR, scanning and trace, — Different filters have also been developed to convert the text and table data into valid XML files. Once the data is available in valid XML format and CGM format, it will then be saved in a repository.
 

Structuring
 
The comparative analysis of the possible solutions has determined the choice of an in-house structuring based on our own DTD. The idea while developing the DTD was to keep as close as possible to the CALS standard to allow for easy conversion between structures.
 
None of the conversion has been successfully automated 100%, and the key factor still remains the human eye. The produced structured data has to be submitted to review by the technical authors.
 
The aim has been to develop processes that avoid confronting the technical author with the complexity related to the structure of the documents. It has been decided to use normal word processing systems and electronic publishing systems enabling valid XML usage. These tools must guide the technical writers by presenting them a choice of valid "next element(s)" and asking for attribute values. Other considerations have been ease of use, and the accuracy in detecting and indicating errors during the learning phase.
 

Data repository
 
The major activity is currently concentrated around the data repository system. The functions this system must handle relate not only to document storage, but must also:
  •  keep track of the changes,
  •  manage the granularity of the data,
  •  enable the reconstruction of any published data.
 
As a first step, it has been decided to use a tree structure based on the operating system directories. The second step being the setup of a Web using the same directory structure but facilitating the retrieval of the information with advance search procedures. The third and final step will be the use of an XML repository based on a database approach. The repository is particularly helpful in keeping track of the modifications made to the documentation, and the use of XML enables a simple split of the different components. This situation enables the technical author to modify the information at a single place and considerably decreases the risks of errors.
 
As soon as the data are available in the repository, in both XML and CGM, a set of information (manuals) related to a radar site can be produced by assembling the relevant information and extracting it from the repository. Illustrated Parts Breakdown information can be produced via a database extraction and a list of the related items created with an XML structure. In the future, when the IETM's will not only be published on CD's, but will also be directly accessible from a Web environment, the repository will have to manage additional data and functions related to information exchange. For example the reporting and sharing of preventive and corrective maintenance data. This information will have to be integrated automatically into the IETM at each viewing session.
 

Viewing environment
 
The XML/CGM files generated from the repository are ready for display using an Internet Web browser supporting XSL. The use of XSL enables the display of different views of the same data. This feature is interesting for presenting data for a specific skill level, or for a specific site or sites.
 
Using XML/CGM with XSL improves the portability and is a major advantage for a fast and accurate dissemination of the information. All modern tools such as the Web, CD-ROM and DVD can be used for dissemination.
 
Currently an intermediate solution is necessary which forces an HTML conversion. This is due to the non-availability of freeware for simultaneous use of XML, XSL formatting, Xlink and Xpointers. The current investigations have demonstrated that for the CD version the best solution is to supply the HTML version only, while for the Web version an "on the spot" conversion is possible using the servlet technology.
 

CONCLUSION
 
Although the approach to IETMs outlined in this paper provides for an alternative to the established military standards, this alternative is nonetheless based upon solid CALS foundations; SGML (or valid XML) and CGM for object data, integrated with relational data elements aligned to MIL-STD-1388 and AECMA 2000M. Current progress indicates that the overall approach provides a cost effective, workable solution when contracting for IETMs. The effectiveness in limiting the proliferation of authoring systems within the documentation maintenance environment remains to be seen.
 
An important value-added factor is the experience that accumulates from the hands-on implementation activities. The experience is usefully employed when generating contract requirements, evaluating bids, making decisions on options, and when controlling and testing deliverable products.
Groves For Applications   Table of contents   Indexes   XML-Enabling Enterprise Databases to Simplify Internet Applications