Microdocs, Birthrights, and Pottage Messes   Table of contents   Indexes   Beyond DTDs: constraining data content
 
 

From Stone Age to Electronic Age for Aircraft Technical documentation
 
Bruno   Pouilleau
  Business Manager
  Jouve S.I.
11 Boulevard de Sebastopol
BP 2734
Paris   France  75027
Phone: 331 4476 8613
Fax: 331 4476 8610
Email: bpouilleau@jouve.fr
 
Biographical notice:
 
Bruno Pouilleau
 
1990
 
Postgraduate degree in applied mathematics ( Fluid mechanics ).
1992
 
First year Phd program in Marketing at London Business School.
1992-95
 
Bull : Pre-sales engineer in Information Systems Architecture, Business Representative in charge for selling ERP solutions based on BAAN software to industry market.
1995-96
 
Computer Service Company ( IBS-AGI) in charged for selling solution to industry based on BPCS/SSA.
1996 to now
 
Jouve Système d'Information in charge for selling AirGTI product line to airlines and OEM. AirGIT product line includes data conversion, documentation revision service, retrieval system, Task Card management system.
 
Comparative study of the new car market for marketing review ( 91 ).
 
Speaker at AIC safety conference ( Sept. 97 ).
 
Speaker at AACO conference ( March 98 ).
 
ABSTRACT:
 ATA, Air Transport Association 
Aircraft
Consolidation
 Conversion  
Electronic Documentation
Engine
Engineering
 Maintenance  
Operations
Productivity
Retrieval System
Revision
 SGML 
 
Up to recent times, the technical documentation was seen as a rich burden but nonetheless a burden by everyone. It was also analyzed as cost center from the finance point of view with no impact on business. Therefore , it was a quiet world away from the hectic real world. The tools used to perform the task was kept very ‘simple with no technology’. The role of documentation manager was perceived by everyone included himself with respect to the quantity expressed in tons of paper going through his department. To be good at his job, he has to be good in logistics. The microfilm as a media did not really achieve the expected excellence but still has diminished the quantity. In both cases the end-user was just forgotten.
 
We can observe today the side effects of this approach :
 
  • Problem to keep up to date the documentation especially in remote locations,
  • Turnaround for distributing revisions,
  • Storage costs,
  • Information access time,
  • Information exchange,
  • No integration of documentation into business process,...
 
In parallel, the information technology moves very quickly and is perceived as an competitive advantage and an active component of the strategy in some division of airlines such as commercial or finance. At the same time, the competition is getting harsher and the need for more efficiency in every activity of the airlines is business critical. Some studies and real experience have shown that mechanics spend 25% of their time in information research. Thereby , this technology differential puts forward the actual need for new tools around documentation. Most of these needs are addressed by ATA 2100 specifications
 
I restrict the scope of my presentation to the case of a company that does very little re-authoring of its documentation if any. So let's imagine we want to implement a digital documentation system. Since today most of the existing documentation is in paper or in some electronic output file formats the first questions are :
 
How is it possible to convert this data ? What are the end-user functions one can expect ?
 
How is it possible to change a paper/microfilm organization for aircraft maintenance to an organization based on digital technical documentation for revisions ?
 
Since the documentation is rich, complex and bulky (Tens of thousand of pages with cross-references, graphics and other specific) people have invented a lot of specific routines and tips to overcome their information access problems and to save time in order to deal with pressure and to achieve objectives. Therefore, people have habits. But they also feel that time is up now for a change since ever more productivity is imposed and ever more reliability is required. So the next question is :
 
What are the impacts productivity wise and quality wise ?
 
Beyond the last question there is the most important one that is what the ROI is.
 
So I will address here below each of the questions. But first of all, I would like to remind you the key features of ATA 2100 specifications that aim at making possible the data exchange and reuse :
 
  • Text format : SGML
  • Graphic format : CCITT-G4, CGM
  • Database : ODBC, SFQL
  • Network and platforms : Independent
 
Data Conversion :
 
The documentation is structured by ATA specifications. But the key point is that this documentation is actually a set of cross-related manuals plus markup signs for warning, cautions, revised text or graphics, COC (airline's specific additional data), plus temporary revisions.
 
To complete the picture, airlines rely on the content to insure passenger safety and avoid accidents by respecting regulations based on the right use of a good documentation.
 
Therefore, when it comes to the conversion of no-digital original data to digital data, the quality of the content and of the structure is critical.
 
Both are critical because professionals are used to access information according to a ATA numbers (structure ) and some key criteria e.g. Part Number ( content ).
 
The conversion tools must ensure the respect of DTDs and formats. And if necessary it must control the syntax and semantic for the language (e.g. simplified English ). In addition, the conversion tools must generate the creation of numerous hypertext links for which we have to insure there is no error. So this is clear that simple tools can not provide such a result. The first conclusion is that the conversion process has to be based on SGML from the very beginning especially for hard conversion ( paper conversion ). The reason is that if the tools are based on SGML then one can perfectly control the structure definition. The conversion process implies the use of scanning and OCR techniques. But, the simple use of both will lead to a very low quality level on both structure and content. To ensure a reliable high quality level on content and structure, you have to combine many techniques in order to manage errors. In this case, it means that the service you buy must be phased on expert system in order to manage errors from several OCRs and coordinates of any information within the page. It assists the operators in avoiding structure errors and in helping bad word corrections. It must ensure also the identification of any information at the character level since you have to enrich the documentation by adding hypertext references. So, one of the key features is to segment the pages in a intelligent manner to insure not only a good browsing but also to permit the management of key data such as effectivity, ATA numbering, figure, item , vendor code, part number,....
 
You can also at this step insert hyperlinks based on the recognition of meaningful structures such as NHA ( Next Higher Assembly ) or cross-references to other manuals or to other chapters within the same manual. The output quality of whole system must independent from the volume.
 
So you have now a set of digital documentation. But you have to provide next revisions to your customer. Therefore, you must run a powerful and flexible document management database. It must permit to incorporate only the new or revised pages. It must control configuration. This database must insure a quick turnaround time for the revision service to comply with recommendations of regulatory authorities. The system must also allow the incorporation of company specific information such as engineering orders or other customized information.
 
Once you have the files the appropriate format for the documentation you have to generate the database of validated manuals. During this indexing, you have insure the last quality controls to insure you did not loose any input information and that you have generated all the expected information such TOC (Table Of Content) or number of chapter, ....
 
At this point, one can advise airlines to subcontract all of this process.
 
Ready for the distribution, your browser has to be configured to reflect some key information such as revised text , COC ( Customer Originated Change), TR ( Temporary Revision), Warning, Cautions,.. with specific background and foreground colors.
 
Your browser needs also to configurable to manage the windows according to the user profile (mechanics, engineer, support engineer,...). The end-users will be able to attach multi-formats notes as public or private notes.
 
The aircraft technical documentation represents a huge volume that is revised not only on a regular basis but also in between main revisions. The whole solution has also to provide temporary information revision service to incorporate urgent and critical data such as these reflected by AD ( Airworthiness Directive) and SB ( Service Bulletin ). So this time you must be able to provide to the airline a tool to add on the fly critical information to the current version of the documentation. This Temporary Revision ( TR ) Integrator must accept several formats and permit to manage the validity of any TR against the content of the documentation version. So the paper organization will be affected because of the media ( you can put up to 50,000 pages on a single CDWORM) , of the TR Integrator ( quicker and more reliable to insert TRs into manuals than any organizational procedures ). The effect on paper organization will be even greater if you are networked. In this case a one minute action will replace days of distribution. The library will not be any longer the place where the information is. PC will replace microfilm readers and information will be more easily re-use. The information will be everywhere it is needed for no additional cost since most of the people who use the documentation have already a PC. The airline will be able to retain experience not only by putting video or digital photo to help performing difficult tasks but also by allowing the workforce to attach private notes . There will be more interaction between engineering and maintenance through the use of public bookmarks. Engineering orders and task cards creation will be easy by using export function for text and graphics.
 
Is that all ? not really ! Actually, this documentation contains data related to stock inventory management, to planning , .... If open systems i.e. having API you can integrate the documentation with third party application such as ERP ( Enterprise Resource Planning) or with task card management system. Then in this case , a buyer in the procurement department will be able to dynamically access the IPC to check out information about vendors or parts. Then a planner will have the possibility to access dynamically the referenced manuals of a given task card and interact with the manuals to create a task card.
 
What are the impacts productivity wise and quality wise ?
 
Productivity savings are made thanks to :
 
  • a dramatic retrieval time reduction between 30% and 70%,
  • an improved usefulness of data since you can access data by several ways ( ATA reference, Part number, Error message, Fault Code Number, ....),
  • a greater accuracy of the data ( TRs , Version control , less maintenance rework),
  • Duplication and distribution costs,
  • an integration with maintenance management systems and authoring systems.
 
Quality improvements are drawn from productivity improvements that deal with data accuracy and with better troubleshooting.
 
You will generate greater revenues and savings not only because you will perform a faster turn-around time for aircraft in maintenance that means more aircraft in service, better on-time performance and lower operating costs but also because you will improve flight safety and quality.
 
The typical payback i.e. ROI is achieved within 16 months.
Microdocs, Birthrights, and Pottage Messes   Table of contents   Indexes   Beyond DTDs: constraining data content