This html article is produced from an uncorrected text file through optical character recognition. Prior to 1940 articles all text has been corrected, but from 1940 to the present most still remain uncorrected. Artifacts of the scans are misspellings, out-of-context footnotes and sidebars, and other inconsistencies. Adjacent to each text file is a PDF of the article, which accurately and fully conveys the content as it appeared in the issue. The uncorrected text files have been included to enhance the searchability of our content, on our site and in search engines, for our membership, the research community and media organizations. We are working now to provide clean text files for the entire collection.
The Navy supply system's aging general-purpose computers are overburdened and increasingly less capable of coping with the burgeoning demands made on them. Often, the fleet tends to overlook the critical importance of its own supply system. But if that system should suddenly break down because its computers no longer can handle the work load, every line officer on board a ship will soon take note of the effects. The flow of ammunition, spare parts—any type of material support—will grind to a halt, and the Navy's operational capability will be paralyzed.
Like most large industries, the Navy uses computers to perform a wide variety of functions. Indeed, most any function associated with logistic support is automated to some degree. Of course, computers do not process independently. People review various computer-generated actions and intervene whenever necessary. Nevertheless, the computer has freed a large number of personnel to make discretionary decisions rather than to perform mundane and repetitive tasks.
Most supply operations are automated on a computer. And, as the scope and volume of automated supply functions have increased, the practicality and usefulness of manual backup systems have waned. Although manual backup systems exist, these procedures can make little headway against the large number of critical supply actions which must be processed when a computer failure occurs. The conclusion is evident: Computers and the associated software which support the Navy's supply system must be reliable.
Yet all computers fail at one time or another, and the Navy’s are no exception. They have failed on many occasions because of hardware breakdowns or software errors. Most failures are fixed relatively quickly and easily; others take longer and require specialists. Isolated failures of a major magnitude have occurred and will continue to occur, but these failures are usually corrected before logistic support is severely degraded. In all, the Navy’s record of maintaining its general business computers has been excellent.
The real danger facing the supply system today comes not from the isolated failure of one or several computers, but rather from the wholesale obsolescence of two suites of computers which process the vast majority of the supply system’s work. If anything could be called the Achilles’ heel of the supply system, it would have to be these aging, technically out-of-date, and overburdened computers which keep the supply system functioning. Although these computers are scheduled to be replaced, it will be 1990 before the transition is complete. And though these computers have held up well, one can hardly see how they can cope with the requirements of this decade. The supply system’s computers should be replaced more quickly than current plans envision. Unfortunately, federal computer procurement regulations will not allow this to happen. And, even when replaced, there is no guarantee the new computers will provide adequate supply support. The solution is not that simple; other issues must be addressed and resolved by the Navy.
Although these revelations may surprise the weapon system operators who depend on the supply system, the news comes as no surprise to those who program, operate, and maintain the Univac 494 computers at the Navy’s inventory control points (ICPs) and the Burroughs Medium Systems at the Navy's stock points. The implications of these revelations are quite serious; the Navy’s supply system is heavily dependent on these computers. Without ade
quate automated data processing (ADP) support, the billions of dollars being spent to buy new weapon systems and support those already in the inventory may be providing less defense than we are led to believe we are receiving.
The Navy’s association with computers for general business purposes began in the early 1950s. As civilian industry did, the Navy first automated logistic and financial functions because they were simple, routine, and repetitive. During the late 1950s. various supply activities experimented with computers for local purposes. It was not until the early 1960s. in response to Department of Defense mandated programs, that the first real attempts were made by the Navy to develop standardized programs which could support a broad range of logistic and financial functions. The Navy's initial efforts resulted in two distinct systems: The Uniform Automated Data Processing System for Stock Points (UADPS-SP) and the Uniform Automated Data Processing System for Inventory Control Points (UADPS-ICP) or the Uniform Inventory Control Programs (UICP) system. The two systems reflected the Navy’s wholesale/retail approach to providing logistic support—a philosophy which is still in use when considering the naval community's logistic needs.
The wholesale/retail split among material relates to the purpose of inventory. Wholesale materials are items not designated for a specific activity's use but are for eventual use in filling the supply system's requirements. Wholesale material is centrally managed at an ICP by an integrated manager who oversees the material to meet requirements. The Navy's ICPs usually do not stock material; instead they control its distribution to meet worldwide demand.
In addition to its wholesale mission, the Navy’s ICPs provide program and supply support for Navy- designated weapon systems. In this regard, program support includes those functions involved in deciding who. what, when, and how the Navy will provide logistic support for a given weapon system. Supply support refers to the actual procurement, repair, receipt, and issue of all the items that program support decides are needed. The Navy's primary ICPs are the Ships Parts Control Center (SPCC) in Me- chanicsburg. Pennsylvania, and the Aviation Supply Office (ASO) in Philadelphia, Pennsylvania.
Retail materials are basically the items which are not designated as wholesale. They are items destined for essential consumption or use by a particular operating activity and are managed at two levels—intermediate and end user. Stock points—such as naval supply centers, naval supply depots, naval air stations, and some naval shipyards—are all considered intermediate levels for retail support.
The wholesale/retail approach allows flexibility in providing logistic support for the Navy. The wholesale system allows worldwide management and visibility of common use and highly critical materials, while the retail system provides support tailored to geographical areas and specific types of activities.
The automated UICP system, developed to support ICP activities, was conceived in 1961 and became operational in 1965. Since then, the UICP system has grown and changed as ICPs have been consolidated and operational requirements have varied. Today, the UICP system provides the SPCC and ASO with the ability to perform their wholesale, program, and supply support missions. The original computers used to support the UICP system were four Univac-490s. Through a series of modifications, upgrades, and augmentations (the last major one in 1974), the computers used to support UICP now consist of eight Univac-494s. Although the hardware and software now used at the ICPs differ significantly from the original configurations, a U-494 is basically a bigger and somewhat more capable U-490. In other words, the computers used today at the Navy’s ICPs are based on 1965 technology. After reviewing the many advances made in ADP technology since the mid-1960s, it is not difficult to understand why computer operators, programmers.
and maintenance personnel consider the Univac-494s to be the computerized equivalent of prehistoric creatures.
The situation at the Navy’s stock points, while better from a technological viewpoint, is not much more encouraging. UADPS-SP, developed to support the Navy’s retail operations at stock points, was first processed on an IBM-1410 computer in 1962. As it did at the ICPs, both the hardware and software supporting UADPS-SP evolved until the late 1960s, when it became evident the installed IBM system would not be able to handle the predicted future workload. This was in spite of the addition of IBM-7440 slave computers and IBM-1050 remote terminals at larger sites in 1966-67. In 1972, the Navy acquired Burroughs-3500 computers and implemented new software to support UADPS-SP at the stock points. The new software was named MARK II. and it represented a quantum leap in the ability of the stock point computers to provide ADP support. Primarily because of this upgrade, the stock points now are in a better technological condition than the ICPs. However, the stock points are not well off because their computers service a larger and more diverse audience than the ICPs'.
Except for the major hardware/soflware upgrade accomplished at the stock points in 1972, ADP growth at the ICPs and stock points has been of gradual, incremental enhancements combined with the procurement of an additional number of similarly configured systems. This policy for managing ADP could be called incrementalism. The prime manifestations of this policy are the Univac-494s located at the ICPs. For example, computer main (core) memory gradually grew from 32,000 words in the original Univac-490s to 262,000 words in the current Univac- 494s. Newer and better disc and tape drives added over the years increased available secondary storage memory. In addition. Navy “unique" and Uni- vac-developed enhancements were made to the software. While necessary and successful, each of these incremental changes only postponed the inevitable. The Univac-494s are now stretched to their technological limits. There are no more quick-fix enhancements available for exploitation.
A similar situation exists at the stock points. Here, however, the Navy has had the “luxury" of moving up the family line of the Burroughs Medium System computers. Since 1972, the Navy has moved from the B-3500, through the B-3700 and B-4700, to Burroughs' largest medium system computer, the B-4800. As in the ICP world, the Navy also increased main and secondary memory capacity, purchased more and faster peripherals, and enhanced software.
In spite of these efforts, the Navy’s supply system managers are in a continuous position of searching for adequate computer capacity and capability.
No industry has lowered the cost and increased the capability of its product as the computer industry has done over the past ten years. The Navy desperately needs these more capable computers.
Computer processing methods have changed drastically since the supply system last purchased an entire line of new computers. The trend has been to shift away from large mainframe computers toward smaller computers (minicomputers), which are often connected by local and national networks. The Navy-owned Burroughs and Univac computers are large mainframes which are heavily dependent on computer card input, regularly scheduled production runs (batch updates), and printed output. They have little true interactive capability which allows a user at a cathode ray tube (CRT) terminal to converse with a computer to query, update, and modify data. With a good interactive capability, a user gets up-to-date, specific information within seconds of a request, in contrast to the dated and sometimes voluminous information provided by regularly scheduled updates on printed reports. This new method of processing enhances user productivity and provides capabilities unheard of ten years ago. Interactive computer systems and CRT terminals paved the way for word-processing systems that are improving office productivity throughout the country.
In addition to interactive capabilities, an amazing array of new software allows better use of the computer’s capabilities. New programming languages facilitate the development of logistic, financial, and other end-user programs. Programming aids exist which indicate when a programmer has made an error and even suggest possible ways to correct it.
New data management techniques have also been developed. For years, a single computer application (set of programs) managed its own data exclusive of all other applications on the computer. Today, data base management systems (DBMs) organize data for the programmer with less redundancy and more consistency throughout the system.
Distributed Data Processing (DDP) is another new technique that provides better support and increased productivity. The distribution of data processing has made computer hardware less vulnerable to fire, flood, sabotage, and failure. Computer networks can shift the workload from one computer to another as the situation dictates. Networks can also facilitate real-time exchange of data and information without resorting to such manual techniques as the telephone or message traffic. A local computer network, which connects users to any one of several computers at a specific site through a single CRT, is another emerging concept.
All of these sorely needed capabilities are generally unavailable to Navy supply personnel.
These capabilities are not a panacea which will solve the Navy's ADP problems—they are tools. Putting these tools to use will require individuals who truly understand the supply system and are dedicated to improving it. Our selected personnel are operating and maintaining hardware and software which are but a few years away from the junkyard.
Responding to the demands of the users in the field and acting on its own initiative, the Navy hierarchy has instituted several projects which use minicomputers to provide specialized support. The intent of most of these projects is to mate minicomputers with either the Burroughs or Univac computers to provide additional capacity and improved capabilities. Plans are being made to automate warehouses, integrate disbursing and accounting functions, automate the procurement process, improve material receipt procedures, and develop other end- user applications.
Each of these efforts only attacks the fringe of the problem. These new projects are a form of incrementalism which take advantage of the unique capabilities of the minicomputer and allow the Navy to make some progress with its current supply problems. Since most of these projects have the need to interface to either the U ADPS-SP or UICP systems, they indirectly put a greater strain on the Burroughs or Univac computers. The first step in the solution of today’s ADP problems must be the wholesale replacement of all the Burroughs and Univac computers—a step the Navy recognizes and is now executing plans to accomplish.
Incrementalism, while probably not the optimum way to approach ADP-related problems, has allowed the supply system to grow and, in many cases, provide better logistic support. Both the UADPS- SP and UICP systems (and their associated computer hardware) have adapted extremely well over the years. Both systems have been unqualified successes. In retrospect, the policy of incrementalism used by the Navy was probably the only rational course of action which could produce results quickly for the Navy, as well as all other government agencies, is a prisoner of the ADP procurement process.
The acquisition of general purpose automatic data processing equipment (ADPE) within the Department of the Navy is a time-consuming, difficult, and frustrating evolution. The complexity of this process stems from three major considerations. First, several government agencies are involved in the process. Second, the generally large amounts of money involved in the acquisition of ADPE have generated high visibility and interest, and led to the establishment of numerous rules and regulations to ensure wise management and tight control of funds. Finally, unlike other Department of Defense (DoD) procurements, ADPE acquisitions require an approval cycle before the procurement cycle can begin. ADPE is the only commodity for which DoD does not have unlimited procurement authority. This anomaly came
about as a result of the Brooks Bill (Public Law 89306) which was passed by Congress in 1965. The purpose of the law was to “ ... provide for the economic and efficient purchase, lease, maintenance, operation, and utilization of ADPE by federal departments and agencies.” The law made the General Services Administration (GSA) the sole agency to acquire, operate, fund, and dispose of ADPE for the federal government. It also tasked the National Bureau of Standards (NBS) to develop ADP standards and the Office of Management and Budget (OMB) to develop policy guidance. Individual agencies were to develop their own ADPE requirements, • specifications, and configurations.
The Brooks Bill established the foundation from which the GSA, OMB, DoD, and the Navy have structured their own organizations for the procurement of ADPE. Each agency in this chain has developed its own guidelines and regulations to fulfill the intent of the law and higher authority’s requirements. Frustrating, to say the least.
The long and complicated process which has evolved from the Brooks Bill has served to lengthen the time it takes to procure ADPE. Several examples of this process are afforded by the three Navy
initiatives currently under way to replace and enhance the Burroughs and Univac computers. The Stock Point Computer Replacement Project began in 1980. Projections indicate that it will be 1988 before the new computer hardware is procured, and 1990 (or later) before the first site is installed. The ICP Resolicitation Project (ResPro) began in 1977 and does not project procurement of hardware until late 1982. Initial implementation is not slated until 1983, and it will be another five years before all the present software can be converted to run on the new computer. An even smaller project, dedicated to procuring minicomputers to bulwark the Burroughs Medium Systems, is taking five years from inception to first-site installation.
Today’s ADPE procurement process is an anachronism; the federal government is being regulated by a law which was passed when hardware constituted approximately 80% of a computer system’s cost, and software design, development, support, and maintenance made up the remaining 20%. The law may have made sense when it was passed (many contend it was enacted to ensure that IBM’s share of the federal market decreased), but events have overtaken it and the situation has changed drastically. Hardware costs have been declining for years and now account for about 20% of a computer system’s total cost.
To justify in excruciating detail the requirement for new computers and then to push the justification through a complex regulatory maze for approval make little sense. Granted, controls should be in place; but agencies and individual commands should be given much more latitude. Most of the supply system’s costs for ADP support are in personnel. People must be given adequate tools so they can develop the new computer applications necessary for the Navy in the future. As this self-imposed straitjacket becomes better recognized as the major ADP problem faced by the federal government, pressure should be brought to bear to change the law or. at the minimum, loosen the regulatory shackles. Without a streamlining of the ADP procurement process, the Navy will continually be years behind the technological advances made by the computer industry and will be forced to use a policy of incrementalism. The recently passed Warner Amendment, which exempts certain military ADP systems from the Brooks Bill approval process, is a step in the right direction. But supply and financial ADP systems are not among the exemptions. The issue is bigger than the Navy, but it should be recognized as the cause of today's ADP obsolescence within the supply system. The Navy’s predicament can provide a convincing case study for those seeking to streamline the ADPE procurement process.
Still, even if the regulatory environment is altered, several issues remain to be resolved by the Navy if the new hardware and software being procured is to be used most effectively.
One of the most difficult problems confronting the ADP community within the supply system is the need to further develop policy and plans for the future. As the Navy moves forward with its efforts to procure needed hardware and software, with projects such as ResPro and stock point replacement, it is becoming increasingly more evident how difficult it will be to weave these and other new computer suites into a logistic and financial system which can cleanly and efficiently interface all necessary components.
The quickening pace of technological change— combined with the slow pace of federal procurement—has made ADP planning a nightmare for some of the most knowledgeable and dedicated individuals in the supply system. Also complicating the planning function are the number and diversity of new demands placed on the supply system. For example, because of the uniqueness of design of many of its computer applications, the Navy has experienced problems integrating new requirements with existing and proposed systems and responding to DoD-mandated programs in a timely fashion. An especially troublesome area has been the interfaces between different computer applications, which seldom mesh neatly.
Planning, especially difficult in the ADP world, is receiving increased emphasis today and will continue to receive top priority. The task of planning the migration from today’s systems to those of the future will continue to be a challenge requiring close coordination between many organizations and individuals. Planning and ADP procurements must be closely linked; if the procurement process speeds up appreciably as a result of President Reagan’s and Congress’s deregulation initiatives, the Navy will find itself in an especially difficult position.
A second issue relates to the increasing pace at which computer technology is advancing. The Navy should devise some type of contractual vehicle which allows a vendor to technologically refresh computer equipment during its life cycle, or the Navy could find its new computers obsolescent within two to three years after procurement. Incrementalism has forced the Navy to develop nonstandard computer systems which vendors find difficult, if not impossible, to adequately support. Once again, there are initiatives under way to incorporate this ability into contracts; however, it has yet to occur. A policy which allows or even forces vendors to maintain, upgrade, and replace Navy-owned equipment over the life cycle of a computer system would be a major step in the right direction.
A third issue concerns computer redundancy. As the Navy continues to become more dependent upon computers, failures will become increasingly less bearable. In the past, GSA has not allowed computer resources to be configured in either a “hot” or “cold” standby condition in order to provide automated backup services because these configurations were considered a waste of resources. A new approach to ADP redundancy is computers capable of backing themselves. This approach uses all computer resources during normal operations. When a computer hardware or software failure occurs, system operation and response time degrade, but all computer functions remain available. This approach is more expensive than a single “regular” computer (because of built-in redundancy), but is necessary in today’s environment. Maybe this new approach will find favor with GSA. In any case, the Navy’s computers must be more reliable and have access to some type of automated backup. ADP is no longer an enhancement to a manual supply system: ADP is the system. Without ADP, logistic support becomes extremely inefficient. Redundant computer systems, no matter the form, are not a waste of computer resources; they are relatively cheap investments which can ensure that our weapon systems remain ready for use.
A last major issue the Navy must confront concerns its heavy investment in existing software. All computer software is greatly dependent upon the hardware which it operates. No program is entirely transportable from one computer to another. Programs written in higher level languages are much more transportable than those written in assembler language (Assembler languages execute about one computer command per line of code, while higher level languages—generally easier to read and decipher—execute several computer commands per line of code.) Because of its failure to upgrade software from assembler to higher level languages in the late 1960s and early 1970s, the Navy faces a monumental task in converting its software to run on the new computer equipment which will be procured in this decade. Unless the Navy mandates that all software be developed in a higher level language, existing software will be converted from one assembler language to another. This will only perpetuate today’s hard-to-transport software into the future. No computer language is truly transportable; however, the new high-level languages are much more capable and gaining wide acceptance. The Navy should investigate the use of one of these languages (possibly PASCAL) for use until ADA (DoD’s mandated computer language of the future) is ready.
The Navy now finds itself at a watershed in its use of general purpose ADP computers for logistic support. The large mainframe Burroughs and Uni- vac computers, now providing support for the Navy, are in obvious need of replacement. Plans are being executed which will replace these computers during the next ten years with newer mainframes and minicomputers. The next few years present an unparalleled opportunity to provide enhanced and cost- effective logistic support well into the 21st century. The key action is to accelerate the federal automated data processing equipment procurement process. Unless this occurs, the Navy will be forced into a less-than-efficient mode of ADP operations, which may require costly alternative actions in order to provide the logistical support required by the fleet.
Lieutenant Commander Bristow graduated from the U. S. Naval Academy in 1973 and has been a supply officer since 1976. He earned a master's in computer systems management from the Naval Postgraduate School in 1979 and currently is assigned to the Navy Management Systems Supply Office.