The Impacts of Standardized
Information Management
Processes on NDA and NDE
Waste Characterization Systems
By Cecelia J. Lemons, Kiona Scientific, and Kurt Conrad, The Sagebrush Group
Abstract
On one hand, the waste characterization industry involves some of the most sophisticated, state-of-the art science and technologies on the planet, on the other, it is cursed with extremely primitive computing practices. This results in pockets of platform-specific software and data that become barriers to interoperability and integration. From our perspective, this industry has a choice: 1) to continue with current software practices, which are likely to result in ongoing redevelopment, higher costs, and customer dissatisfaction, or 2) to embrace the emerging standards-based information management principles which are beginning to dominate the investment strategies for both government and commercial customers. This decision may be critical because it is not just a choice between competing information technologies but is part of a broader set of information and management policies that offer important benefits to both vendors and customers.
The
Problem:
Waste Characterization Software
The primary challenges to this industry is not hardware evolution but the development of supporting software software systems. Today, the software systems that support NDA/NDE equipment are isolated, platform and process specific, and primative in terms of functionality. In addition, software development in the waste characterization industry is characterized by long development cycles and expensive re-development cycles. The result is pockets of data that can not be easily synthesized to produced needed answers.
The shopping list of unmet needs are well known to many within the industry:
- Methods integration and application interoperability
- Data fusion and data reduction
- The integration of NDA/NDE results with other customer information
- Solid regulatory compliance
Of all of these problems, issues of regulatory compliance are probably the most critical. From our perspective, regulatory compliance will necessitate the integration of NDA/NDE results with a wide variety of process-specific information sets, including:
- Waste acceptance criteria
- Waste management (storage and processing) records
- Operating procedures and calibration processes
- Configuration control documents, including pedigree requirements
- Audit and assessment reports
- Performance demonstration assessments
The problem should not be suprising. Historically, DOE has emphasized the application of these technologies to safeguards and the information management systems have been targeted at those applications. Funding to adapt these technologies to waste applications has been minimal. The result is that the software systems are insufficient to meet the multitude of regulatory requirements.
While many of these problems may seem unsurmountable, especially given contract and funding uncertainties, this industry's customers are driving themselves and their vendors towards common solutions, which are based on the application of standards-based information management requirements.
The
Solution:
Standards-based Information Management
Fundamentally standards-based information mangement involves the defining specific technical requirements for information at various points in its life-cycle. This is being done primarily for policy reasons:
- To ensure interoperability and data integration
- To provide platform independence
- To protect information investments from anticipated harware and software technology changes
Not suprisingly, many of these same policy interests are mirrored within the waste characterization industry. It is for this reason that we believe that it is in the best interests NDA/NDE vendors to agressively embrace these many of these information management standards, in some cases, ahead of explicit regulatory requirements.
Many of the specific standards are being chosen in the context of heterogeneous computing environments, and with the expectation that these multi-vendor information management environment will exist well into the future. Accordingly, most of the standards-definition activity is being focused at the data encoding level. One of the best examples of an encoding standard is that used for CD-ROM disks. The actual physical characteristics of the media and how digital information is represented on that media have been carefully defined. The result is that CD-ROM disks are fundamentally readable on virtually any computer (Apple, IBM, Sun, etc.) that uses a conforming CD-ROM drive and it is not uncommon for software for all three platforms to be shipped on the same disk.
Text Encoding Standards
For many, if not most, of the applications being discussed here, the selection of the text encoding is probably the most critical decision. There are a wide range of text encoding approaches ranging from ASCII text to popular word processing formats and their derivatives (e.g., WordPerfect, RTF, Frame) to encodings designed primarily for the distribution of text (Adobe's PDF format and a number of competitors). Most of these encoding formats, however, suffer from the same limitation, they are controlled and defined by software vendors or groups of vendors. This is what is driving the selection of SGML (the Standard Generalized Markup Language).
SGML is an international standard (ISO 8879) and has a companion standard called HyTime (ISO 10744), which is provides an array of multimedia and hypertext extensions to SGML. SGML and HyTime increase the value of an organization's greatest assets—the knowledge and information in its documents—by making hem accessible and useable across platforms, applications, users, and time.
Because SGML and HyTime are international standards, their definition is controlled by groups of software vendors and customers and can only be changed through a formal balloting process. This helps to assure that changes to the standard will not be done at the whim of a single software vendor and that SGML documents which are produced today will be usable years and decades from now.
Who's Adopting SGML?
The federal governement has adopted SGML as a computing standard (FIPS-Pub 152). Additionally, OMB Circular A-130, Management of Federal Information Resources of June 1993 requires agencies to plan in an integrated manner for managing information through out its life-cycle. It further recognizes the free flow of information between the government and the public is essential to a democratic society. SGML is a critical enabler of this free-flow of information and is being used as the basis for electronic interchange in the DOD, DOE, IRS, Security and Exchange Comission, Patent and Trademark Office, and other federal departments and agencies.
It might appear that the federal government is bombarding free enterprise with more worthless regulations. But a closer look at SGML trends shows that most of the growth is occuring in the high-technology sectors of the U.S. and international economy. Aerospace, semiconductors, software publishing, automotive, and pharmeceuticals are just a few of the sectors that have active SGML efforts, many of them going back for a many years.
SGML Use Within the DOD
Within the federal government, the DOD is the largest single user of SGML. It adopted the standard before international balloting was complete and has developed a number of important applications. One of the most notable ones involves IETMs (Interactive Electronic Technical Manuals) to support the maintenance and operation of sophisticated weapons systems.
We believe that the adoption of DOD's IETM standard (or an appropriate subset) by the waste characterization industry would provide extraordinary benefits in the areas of regulatory compliance and accountability.
SGML Initiatives Within the Nuclear Industry
Although SGML use within the DOE has lagged behind that of the DOD, current initiatives can be traced back to 1991, through memorandum and strategic plans for the electronic exchange of Scientific and Technical Information (STI). These plans were codified as requirements with a revision to the DOE Order 1430.1D, Scientific and Technical Information Management, effective June 30,1994.
Among other things, the STI Order directs that these requirements be contractually administrated and be included in strategic plans, institutional plans, and program guidance, as well as performance measures for contracts, grants, cooperative agreements, and Cooperative Research and Development Agreements (CRADAs) under the DOE's purview. Implementation of this directive must be included in all performance evaluation activities.
Because waste remediation activites are specifically defined as a generator of STI, waste characterization vendors will find their software systems increasingly scrutinized for compliance with the SGML standard. In addition, the Order's life-cycle information management criteria stipulates that information resources be managed as long as the information is useful. For the long-lived radio-isotopes that is truly a long time.
DOE is not the lone adoptee of SGML within the nuclear field. The NRC and a number of utilities have been working on SGML applications for many years. The Electricite de France has launched an Electronic Library Project that is based on just on SGML, but uses HyTime as the primary integration mechanism for tens of thousands of reports, contracts, serials, programs, and videos.
NDA/NDE Industry Benefits
We believe that the future is clear. Participants in the waste clean-up arena will need to ensure that their systems fulfill emerging life-cycle information management requirements, both to meet the customer's need and to protect themselves from legal liability. They will be using SGML to do it. Without standardized approaches, there is no common denominator on which to construct that crucial interface between organizations or meet regulatory requirements.
In addition, we expect that the timely adoption of SGML and HyTime would solve a number of immediate problems and help waste assay software to be fully interoperable across platforms and support the exchange of data, analyses, and reports.
While SGML may be important for designing robust information processing systems, its greatest value within the NDA/NDE industry may be improve organizational adaption and resiliency. The value of these comprehensive and integrated documentation and reporting systems will become even more apparent with the introduction of ISO 1400. ISO 14000 is an international non-regulatory environmental standard that may be required to conduct business between the corporate and regulatory leaders for environmental management as soon as 1996.
We envision a wide range of SGML applications that are defined through the active participation of vendors, government, and commercial customers:
- Regulatory Compliance (in Real-time)
- Integration with upstream and downstream data maangement processes
- Real-time management oversight of large-scale waste processess
- Performance demonstration assessments
- Multi-systems integrations
- Multi-media including multi-analyses reporting
- Online procedures, processes and technical manuals (IETMs)
- Diagnostics and Maintenance of assay systems
- Remote controls
We also envision SGML and HyTime being applied to research activies, such as Monte Carlo, where the quality of the research could be dramatically enhanced by creating large data repositories of input files in open format to reduce the time required to input data. We also see improved information management speeding much of the research and development that is still needed in the areas of hardware and instrumentation.
Any well-engineered software system is expensive, but cost/benefits of using SGML have already been well documented by current customers and the cost of commercial SGML tools is falling rapidly with the introduction of products by leading commercial software vendors (e.g., Microsoft and WordPerfect). Obviously the time to develop and SGML application varies with the complexity of the task, but experience has proven that information modeling with SGML is faster and simpler than developing traditional relational database models.
Conclusion
The biggest challenge today faced by the NDA/NDE industry is to adopt these emerging standards and information management policies in order to cope with the chaos which exists in today's multi-tiered and overlapping regulatory environment. Not only are the information mangement policies and standards described in this paper being demanded and mandated by customers, but we also see significant tactical benfits to be gained by applying these computing standards to the outstanding software problems facing this industry.