Since the original version of this focused program paper was published in 1995, manufacturing has become more "demand driven" as opposed to "plan driven", and individual manufacturing sites are expected to behave with greater autonomy ⁽¹⁾ . This shift in the manufacturing culture forces manufacturing execution systems (MES) software to expand to meet all scheduling and tracking needs within the walls of a plant, and to anticipate requirements, as opposed to schedules, from the upstream enterprise resource planning (ERP) systems ⁽²⁾ . Concurrently, networked object technology has matured sufficiently to encourage industry to work through consortia, such as the Open Applications Group (OAG) and the Object Management Group (OMG), to engineer more of a component approach to MES into industry standards and products. To meet the changing needs of the manufacturing sector, the scope of the Technologies for the Integrations of Manufacturing Applications (TIMA) focused program has been fine-tuned. TIMA is now open to the entire spectrum of manufacturing, both process-oriented and discrete, to encourage a market in re-usable MES components. The program also recognizes the need for MES to integrate even higher into enterprise management software, to ensure that "available to promise" calculations can be based on current factory status.

ABSTRACT

Companies in many manufacturing industries face the challenge of responding more rapidly to changing markets and evolving business opportunities. Today, the speed with which new products are developed and delivered to market often is the chief determinant of competitive success. Moreover, as manufacturers trim cycle times, they reduce production costs, another advantage that can be leveraged in the marketplace, in the form of lower prices and more timely introduction of products.

While U.S. manufacturers recognize the importance of speed and flexibility, many are impeded in efforts to become more agile producers. Even highly automated plants and factories struggle to overcome difficulties in adapting or reconfiguring production operations to accommodate design changes and new product lines. Idiosyncrasies in manufacturing software and incompatibilities among software applications are a primary source of costly delays, because customized systems integration efforts are often needed to solve these problems.

This multi-year focused program within the Advanced Technology Program promotes greater speed and agility among U.S. companies in the manufacturing sector. Technical work carried out under the program will facilitate industry-led efforts to create a real-time, "plug and play" manufacturing software environment. The overall technical goal is to develop and demonstrate the technologies needed to create affordable, integrable manufacturing systems, that is, those that can be rapidly integrated and reconfigured and, ultimately, that can automatically adjust their performance in response to changing conditions and requirements.

The ATP focused program will benefit small and medium-sized companies by enabling them to gain the performance advantages offered by previously unaffordable automation tools and will benefit large manufacturers by enabling them to become more agile. In fostering development of tools needed to create robust, straightforward means of achieving connectivity and interoperability among the varied components of manufacturing execution systems (MES) and the information infrastructure of an enterprise, the ATP focused program will also enable the expansion of the MES market, spur innovation and competition among MES vendors, and by so doing will lower system and software costs to levels within the budgets of smaller businesses.

This program is open to proposals that address any manufacturing industry. In the short term, the utility of the tools and technologies resulting from this program will be greatest in industries in which the life-cycle of products is decreasing while the variety and technical complexity of products is increasing. Similarly, integrable MES will be of significant practical value to companies engaged in distributed manufacturing and to those that are developing close operational links with suppliers and customers.

PROBLEM STATEMENT

The costly and massive effort required to implement and integrate information systems that share real-time manufacturing data throughout the organization is a significant impediment to creating more agile manufacturing businesses and industries. Typically, factory-floor information systems focus on the operation of production equipment and the control of processes. They communicate neither directly nor regularly with front-office information systems or with design and engineering systems. As a result, upstream information systems are unaware of important manufacturing details, such as the availability of appropriate tools, labor, and materials; maintenance schedules; records of past process performance; or the status of work in progress. Middle-level information systems, known as manufacturing execution systems (MES), bridge this critical information gap between upstream and downstream activities. Today's MES solutions, however, are burdened by complexities that make them difficult to implement and integrate and, often, even more difficult to modify and upgrade.

MES applications track and manage all aspects of a job on the shop floor, at any point in the production cycle, in near real time. MES applications play essential supervisory and monitoring roles that link all levels of manufacturing and business operations. For example, they identify bottlenecks and material shortages on the shop floor, and they provide up-to-the-minute process performance results along with comparisons to past performance and to projected business results. While these tasks are important to the management and operation of all manufacturing businesses, only a small fraction can afford the cost of installing and maintaining an integrated MES solution, estimated to run between $400,000 and $1 million ⁽³⁾. The sizable costs incurred for maintenance and systems integration work required to upgrade or otherwise change the system place MES further out of the reach of smaller manufacturers. On average, a factory with an installed system spends $2 for in-house integration, maintenance, and support for every $1 expended for products and services sold by MES vendors ⁽⁴⁾. Consequently, only larger manufacturers with average annual revenues on the order of $50 million or more are currently adopting integrated MES solutions.

Across discrete manufacturing industries, adoption of MES has been slight and uneven. Large companies in industries with substantial regulatory and reporting requirements, such as the pharmaceutical, aircraft, and defense industries, are the most common users of MES, followed by semiconductor manufacturers, which manage complex, hard-to-control production and assembly processes.

Three approaches to implementing MES have been followed by the small number of businesses that have invested in such systems:

Proprietary, in-house solutions--developed internally or by engaging the services of systems integrators that build customized systems.

Point solutions--fashioned from commercially available, individual applications that are then custom-integrated with common database and messaging systems.

Commercially available, integrated MES solutions--an integrated suite of interoperable applications but requiring customized integration to adapt these applications to specific needs or to implement new applications and capabilities.

Each approach has its merits, but all share the same fundamental shortcomings that add significantly to the costs of a system over its lifetime and ultimately undermine manufacturers' efforts to become more agile. That is, all three approaches are inflexible to change, and none of the approaches can be seamlessly integrated into the enterprise--major deficiencies in a rapidly changing, competitive environment and during a period of accelerating innovation in technology and in manufacturing and management strategies.

BUSINESS GOALS (5-10 YEARS)

The ATP focused program on Technologies for the Integration of Manufacturing Applications (TIMA) facilitates a fourth approach--integrable MES--that achieves the benefits of the three existing approaches while enabling important new ones. Like point solutions, an integrable MES can be assembled from piece-like and reusable components. And, like proprietary solutions and commercially available suites of applications, integrable MES provides comprehensive, system-wide monitoring and reporting. But, unlike existing approaches, integrable MES will be more comprehensive, and will accommodate rapid customization, incremental installation, as-needed reconfiguration, distributed monitoring and control, and enhanced information flow throughout the enterprise.

Anticipated outcomes of the capabilities enabled by integrable MES and to be realized by companies of all sizes across a wide range manufacturing industries include:

• Significant reductions in manufacturing costs and cycle time, enabling cost-effective production of both large and small lot sizes and customized products. For example, engine transfer lines have had as much as 50% down-time due to their complexity. MES production monitoring systems have been shown to increase up-time by more than 20%--a solution which is much less expensive than installing additional machines.
• Performance advantages offered by previously unaffordable automation tools. Small and medium-sized companies especially, but other companies as well, will benefit by enabling them to gain the access to these new tools.
• Significant reductions in time-to-market for new products. Current integrated MES solutions can take over a year to install in larger manufacturing operations. Integrable MES should require little more than the three weeks it now takes to install most point MES solutions. This can be achieved through incremental installation at very low systems integration costs.
• Reductions in the time and cost of starting up new factories and of changing over or recommissioning existing factories, which would help smooth fluctuations in manufacturing employment. For a typical 12-month life cycle product, being one month late to market can forfeit 50% of the available profit. By adopting interoperability technology at the equipment control level, one major manufacturer realized a 5X reduction in equipment integration cycle time.
• Reductions in inventory and higher levels of capacity and resource utilization. The agility offered by being able to plug together best-of-breed applications will allow an optimal match between software systems and current factory requirements.
• Reductions in the time and cost of establishing and operating tightly coupled yet flexible supply chains, enabling direct communication between MES systems across company boundaries.

While work carried out under this focused program will be limited to the development and demonstration of early-stage integrable MES technologies and methods, subsequent refinement and commercialization of these technical outcomes by industry should result in significant impacts within five years of the program's completion.

TECHNICAL GOALS (5-10 YEARS)

This program's technical goals are to research, develop, and validate the technologies, methods, and infrastructure needed to reduce the cost and time required to integrate manufacturing execution systems into a manufacturing enterprise. This may be accomplished through a combination of infrastructure and the development of easily integrable applications. This program intends to ease the flow of real-time manufacturing execution data among MES applications, as well as to make real-time data available upstream to executive information systems (EIS) and enterprise resource planning (ERP) systems, laterally to design systems, and downstream to equipment control systems.

TABLE 1. Sample research topics supporting technical goal characteristics.

Integrability	Configurability	Adaptability	Extensibility	Reliability
Frameworks Integrated modeling & simulation Embedded integration technology	Legacy system integration Concurrent information technology paradigms System emulation Model-driven MES Development environments	Goal-seeking, autonomous manufacturing technology Self-organizing, self-configuring, self-correcting manufacturing applications Techniques to map local goals to global behavior	Scability of solutions Networked business objects Distributed object services	Assessment and tuning of complex, distributed, real-time networks Maintenance and migration Security System and data integrity and consistency

The research will support a real-time manufacturing environment which is rapidly and easily reconfigurable, promoting increased agility, and exhibits the following characteristics:

• Integrability of manufacturing software across platforms and vendors' product lines.
• Configurability of a factory's manufacturing system architecture to allow for rapid introduction of new products, and rapid response to unanticipated changes in requirements.
• Adaptability of a factory's manufacturing system architecture to automatically adjust to changing work loads and conditions.
• Extensibility of solutions to complex manufacturing requirements within single factories, to other sites, and to other phases of the product life cycle.
• Reliability of distributed systems, and integrity of shared data.

Technology Barriers

There are numerous technical barriers, some owing to the complex nature of real-time data, which must be overcome to achieve manufacturing execution system interoperability in a general and reusable manner. The following lists some of the unresolved issues:

• The automation of legacy system integration is not well understood, and requires techniques to map information and behavior across different technology generations and system architectures. The real-time nature of MES data further complicates the integration of manufacturing data within an enterprise.
• A certain degree of functional tailoring will always be necessary to meet the needs of a specific, changing manufacturing facility. No formal, widely applicable methodologies exist.
• Critical production operations, particularly those in multi-site, multi-national enterprises, require systems that can be modified on-line, in real-time. Current technology is insufficient to meet this demand.
• Adaptable systems require new ground to be covered in the area of autonomous software systems. Research has not matured in areas such as agent technology, self-describing interfaces, support for multiple communication paradigms, knowledge representation, sharing and reuse, capture of knowledge and context, information mining, and self-modifying software entities.
• Technologies to support self-integrating applications are currently non-existent. This is similar in concept to installation programs for personal computer applications that query the characteristics of the computing environment to customize the installation.

This program will support research which addresses technical goals and barriers described above, including, but not limited to the following:

• Development of open manufacturing software infrastructure, and demonstrating its impact in compliant applications.
• Embedding advanced technology within MES applications for ease of integration with both legacy and future systems, and development of mechanisms to concurrently support various information technology paradigms.
• Integration of simulation-based tools via a common simulation infrastructure.
• Robust methodologies to characterize manufacturing activities, data, business processes and communications in a consistent manner.
• Development of MES application development environments, including application of simulation, emulation and virtual reality.
• Development of model-driven MES software applications.
• Development of self-organizing, self-configuring, self-correcting manufacturing systems.
• Development of techniques to map local goals to global behavior and to a distributed, autonomous software architecture.
• Demonstration of viability of a distributed, goal-seeking, autonomous software paradigm for large-scale, manufacturing operations.
• Scaling of solutions to all sizes of manufacturing facilities, and to all phases of the product life-cycle.
• Development of technology to support finding, accessing, and advertising the availability of object services.
• Development of metrics to assess the availability and organizational impact of distributed, real-time manufacturing system architectures, and mechanisms to tune system performance and ensure data integrity.
• Development of appropriate ways to address maintenance and migration of systems which do not require total shutdown of the operation.

This five-year program started in 1995 and is expected run through 1999. This is the second of three competitions planned at approximately yearly intervals. The total budget is estimated to be $120 million, although funding for the third competition is contingent on future ATP appropriations. Private-sector participants must meet ATP matching-fund and indirect-cost requirements.

Like all ATP programs, this focused program will provide cost sharing only for technically sound projects aiming to develop high-risk technologies with the potential for broadly based benefits. It will not fund consensus-building or standardization efforts, but it can provide support for critical research and development that underpins the creation and definition of standards. Proposed R&D must be clearly innovative and challenging, in terms of either component technical risk or system risk. Software agents applied to link MES components and other systems are an example of a technology that confronts system risk. It has been demonstrated for small-scale applications, but it has not been demonstrated across an entire system or in a realistic production environment.

This program's objective is to develop interoperability solutions applicable to all manufacturing sectors. This represents an enlargement of the scope since the 1995 TIMA competition. The scope now includes:

• Technology, including, but not limited to: agents, objects, request brokers, relational and object-oriented database technology.
• Breadth of proposals: from single shops to multiple sites in an enterprise.
• Types of software: applications, infrastructure, tools, and environments.
• Product life cycle: manufacturing execution systems (MES) and production engineering, including planning and scheduling for the shop floor, but excluding product design, distribution, or maintenance.
• Industry sectors: all sectors including automotive, electronic/electrical, aerospace, apparel, and continuous process industries.
• Business hierarchy: factory site or sites through shop floor to work cells, but excluding higher corporate levels and low-level machine elements like sensors.

Exclusions

Note that ATP funds can not be used for work beyond the proof of concept or early prototype stages. The following areas would be considered out of scope:

• Integration of two or more manufacturing applications without employing integration technology that could be transferred to other systems (i.e. point solutions).
• Development of stand-alone MES applications that do not contribute to the overall goal of an adaptable "plug and play" manufacturing environment. For example, advanced stand-alone schedulers based solely on new scheduling algorithms would be considered outside the program's scope.
• Consensus-building activities, including large-scale demonstrations of essentially proven technologies to foster industry endorsement.
• Development of specifications or data models without provision for realistic proof-of-concept demonstrations.
• Electronic commerce proposals unrelated to manufacturing production.
  
  

Manufacturing is the backbone of the economic engine in the United States, from the smallest electronic components to the largest heavy machinery and transportation equipment, and represented over $1.5 trillion in sales in 1992 ⁽⁵⁾. Discrete part manufacturing alone accounted for well over one third of all manufacturing jobs ⁽⁵⁾. Starting with this competition, the focus has been enlarged to include process industries from discrete manufacturing alone. The result is that the government investment will be leveraged into savings totaling in the billions.

A successful ATP technical program, with committed follow-through efforts by the private sector, will benefit companies across a range of industries:

• Companies that already employ MES, by providing them with a wider range of powerful, integrable applications that will dramatically improve the manufacturers' ability to reconfigure, scale and adapt their processes.
• Small and medium-sized manufacturers, by making MES more affordable and by providing a direct path toward greater automation through incremental additions of new, easily integrated applications and capabilities.
• Vendors of MES products, by expanding the market, lowering barriers to entry, and providing incentives for innovation and technical specialization, since new products and services can be marketed to a broader range of potential customers.

For consumers, widespread adoption of integrable MES will translate into greater product variety and greater availability of customized, affordable manufactured products.

Integrable MES has the potential to strengthen the competitive capabilities of large segments of the U.S. manufacturing sector, creating opportunities to build existing market share and to be first with quality products in new markets. Economic benefits would flow to the larger service sector, much of which embodies or uses manufactured goods in the delivery of services.

INDUSTRY COMMITMENT

This program has been developed on the basis of input received from a variety of industrial sources, including current and prospective users of MES, suppliers of MES frameworks and other infrastructural technologies, vendors of MES applications software, and systems integrators.

The American Electronics Association highlights systems integration as one of four critical areas needing R&D resources, while the National Technology Roadmap for Semiconductors lists manufacturing integration tools as one of the industry's highest priorities. Similarly, the National Electronics Manufacturing Initiative (NEMI) includes MES integration among the most important infrastructural issues that must be solved in order for U.S. electronics manufacturers to succeed in the global marketplace. NEMI cites the lack of an agreed-upon open framework for integration as a major impediment to future competitiveness, and identifies ATP focused programs as an important mechanism for meeting this need. Recently, the National Center for Manufacturing Sciences (NCMS) published a report ⁽⁷⁾ which presents the view of, "leading manufacturing industry experts who are well positioned in the industry to understand the forces that will influence North American manufacturing." In the report, the authors state:

...automation still consists largely of technology islands with little site-wide integration focused either on problem areas or on new process portions of the overall plant.....The demand for computer-integrated manufacturing (CIM) software is growing exponentially. Rapid growth in hardware capability increases the perception that major improvements and applications should be possible. Yet software development and deployment methodologies have changed little over the last 30 years. The result is an explosion in demand for integrated CIM applications that cannot be met because available resources are committed to application maintenance....Legacy systems implemented with no intention of ever being integrated originated as solutions to specific problems in a single department without regard to the need for integration. In many cases, the problem can only be solved by replacing legacy systems with systems designed for integration.

This last point is key to this ATP focussed program, which is targeting the creation of new technology to facilitate the integration of both new and legacy manufacturing systems.

On February 27, 1995, an ATP-sponsored workshop was held in Dallas, Texas, to assess the level of interest and commitment by industry. Strong interest was expressed by more than 60 workshop participants, which included: MES vendors, multiple end-user industries (including automotive, electronics, and aerospace), several industry consortia representing more than 500 small, medium, and large companies, systems integrators, and academia. This focused program paper synthesizes the industry input received at the workshop and expressed in about 50 program white focused program papers submitted to the ATP office. In addition, significant interest has been expressed by member companies of several major consortia, including the National Center for Manufacturing Sciences (NCMS), the Manufacturing Execution Systems Association (MESA), Consortium for Advanced Manufacturing-International (CAM-I), and SEMATECH. Some consortia, such as the Object Management Group (OMG) and the Open Application Group (OAG), are directly addressing MES integration within their working groups.

WHY ATP?

The technologies that this focused program aims to foster are primarily infrastructural with proof of concept applications: they constitute an underlying foundation required to enable and support important applications of information technology to manufacturing. Like other types of infrastructure, these sought-after technologies are recognized as being the means to realizing widely shared benefits. Yet, these benefits are often slow to be realized by the originators. This leads to insufficient incentive to develop infrastructure technology on their own. And in the case of manufacturing information infrastructure, individual companies do not have the capabilities needed to develop the full collection of underlying technologies.

MES vendors, for example, are reluctant to invest in the information infrastructure for manufacturing because they are not likely to realize a competitive advantage. The MES industry is too fragmented for any one vendor's integration approach to dominate the marketplace, or for any one vendor to invest substantial resources in the uncertain prospect of establishing its technology as an industry standard. Industry revenues in 1994 totaled an estimated $1.15 billion, for an average of about $4 million per firm. Of the some 280 MES vendors in the United States, fewer than 10 have sales totaling more than $10 million a year. As a result, investments in R&D are modest. If MES vendors allocate 13 percent of revenues for research, as is typical in the software industry, the average MES vendor spends about $530,000 on R&D each year, with the bulk of that amount directed toward new product development. In sum, MES vendors are not likely to develop integrable MES technology on their own.

Individual manufacturers will pay for functional capabilities. They do not invest in infrastructural technology that will yield benefits of equal--and perhaps greater--value to competitors who did not share in the costs. MES vendors realize that, for their industry to thrive, they must develop the basis for architectures, interfaces, and other core technologies essential for truly integrable systems

Playing the role of catalyst, ATP can help MES vendors and manufacturers to overcome this impasse and to focus their collective expertise and to concentrate some of their resources on surmounting the barriers to developing integrable MES architectures and applications. Eliminating these barriers requires the constructive participation of multiple industries to ensure widespread adoption of the technologies spawned by efforts to develop the necessary infrastructure. On the basis of input from a variety of industries, the ATP concludes that both MES vendors and manufacturers see it in their best interests to participate in such efforts. Manufacturers realize that, with today's custom-integrated manufacturing applications, they can not reconfigure their manufacturing operations quickly enough to react to changing markets and new business opportunities. ATP's cost-shared support mitigates some of the "free rider" problem that has stalled efforts to correct the well recognized limitations of current systems. MES vendors realize that, for their industry to thrive, they must follow the lead of the personal computer industry, developing the basis for architectures, interfaces, and other core technologies essential for truly integrable systems.

The ATP's cost sharing leverages industry's limited R&D funds for high risk technology development, and it encourages companies that would not otherwise have the resources to carry out the MES technology development on their own to move out aggressively and ambitiously.

WHY NOW?

There is growing consensus within industry that the current MES approaches will not be able to meet the needs of U.S. manufacturers in global market competition. The installed MES base is still small. Now is the best time to introduce a new implementation paradigm--integrable MES--before the manufacturing sector increases its investment in, and becomes further committed to, inflexible solutions, thereby increasing industry inertia. Not only has industry awareness of the problem grown, but technical solutions have matured beyond the basic research stage. An infusion of ATP funds at this point will result in significant competitive advantages for U.S. manufacturers and high payoffs to the national economy.

REFERENCES

1. MES II for Discrete and Repetitive Manufacturers: Have the Requirements Changed?, Report on Manufacturing, October 1996, p. 11-22.

2. MES II: Networked Business Objects in Plant Operations, Report on Manufacturing, July 1996, p. 3-11.

3. Presentation made at the ATP Manufacturing Workshop, February 27, 1995, Dallas, Texas, by Ed O'Rourke, Manager, Product Marketing, Consilium, Inc.

4. Integratable MES: The Challenge and the Opportunity, proprietary study, Advanced Manufacturing Research Consulting, 1995.

5. Statistical Abstract of the United States, U.S. Department of Commerce, Economics and Statistics Administration, Bureau of the Census, September 1995.

6. Technology, Economic Growth and Employment: New Research from the Department of Commerce, U.S. Department of Commerce, Economics and Statistics Administration, Office of the Chief Economist, December 1994.

7. NCMS Collaborative Manufacturing Agenda, National Center for Manufacturing Sciences, NCMS Document 0040RE96, May 1996.

Date created: February 12, 1997
Last updated: April 12, 2005