Naval Engineering: Alternative Approaches for Organizing Cooperative Research -- Special Report 266 (2002)

All of the cooperative research organizational models considered by the committee possess operational or functional elements that affect their ability to accomplish the goals and objectives of the Office of Naval Research (ONR). In general, the implementation of these elements is independent of the structure of the organization. The success of any model, therefore, will be affected by such factors as how the governing bodies are constituted; what decisions are made on administrative processes and controls; and how decisions are made about personnel, location, and other management attributes. Thus, the committee elected to discuss these elements separately from its evaluation of the three models themselves. In this chapter these key operational elements are presented, and the implications of each are discussed.

A fundamental issue in structuring a cooperative research program to meet ONR’s goals is the process and manner of setting the research agenda. In the context of a cooperative research program, stakeholders [Navy (ONR, Naval Sea Systems Command, Laboratories), academia, industry, and professional societies] all have a shared interest and ownership in the research agenda. Successful organizational models have structures and mechanisms to ensure appropriately balanced representation and input to the research agenda from stakeholders.

A research agenda usually includes various levels of specificity. At the highest level are research themes that define broad areas of research in systems and technologies essential for future naval capability. At the next level may be several specific projects within these themes that incorporate research to further knowledge in a system or field of application. The responsibility for developing the research agenda within a research organization should be shared among stakeholders. It can be established either top-down (ONR-defined) or bottom-up (investigator-defined).

Under the baseline individual investigator model, while general guidance for a research agenda usually comes from ONR, the impetus for specific

projects is generated from proposals coming from individuals or teams of researchers. This places the burden of project review and funding decisions on ONR and thus leads to an ONR-defined agenda. Under the cooperative models, however, most of this effort is usually shifted away from ONR to the host organization and thus leads to shared responsibility for setting an agenda.

In the case of the professional society model, a research council comprising the stakeholders would assume the responsibility for defining research themes. Similarly, under the consortium model, the executive committee, again representing the stakeholders, develops the research themes for ONR’s approval. The solicitation, review, and selection of individual projects under the themes are responsibilities of the executive committee. In the case of the consortium and professional society models, the executive committee would be established and facilitated by the host organization. The project-centered model represents a somewhat different approach in starting up the organization. ONR would need to create an executive council of stakeholders to oversee the process of identifying the annual project theme; then, a technical review committee would be established to review and recommend projects to promote the annual theme. Therefore, the professional society and consortium models generally produce a more coordinated research agenda than the portfolio of individual projects that might be expected under the project-centered model.

The venue and institution selected to host a cooperative research organization would be expected to provide administrative, accounting, and human resource services. Under most models, the fixed facilities would be minimized to provide more operational flexibility. The nature of total ship systems naval engineering is such that it does not require an extensive laboratory infrastructure, so the existence of major facilities would not normally be a major consideration in selecting a location. Since the naval engineering community is very small and diverse, the number of locations from which to choose is small, and there are few logical central locations where many professionals are concentrated. Therefore, the decision to choose one institution among the few available options will be difficult and contentious.

No matter what organizational model is used, the location of the research organization generates exceptional stature for the host entity while reducing the stature of those entities not selected as host. Because of the small size and fragile condition of the community, a reduction in the competitive stature of an institution could endanger one of the schools currently engaged in naval engineering education. To ameliorate this effect, the organization’s committee structure should very seriously consider the location of the research center and the structure of the decision-making groups within the organizational model. The goal would be to create a balance in the influence of participating entities.

A cooperative research organization would have several locations to choose from. Three of the more obvious options include a university, a neutral (nonstakeholder) location, and a federal government facility.

A university location could be either on or off campus. An on-campus location has the advantage that existing infrastructure and administrative staff might be available and thus reduce center start-up time and cost. An offcampus location might require more investment in infrastructure. Both types of location would provide strong support to the existing marine, ocean, or naval engineering department faculty and students at that university. The resulting increased strength and technical capability in naval engineering and related subjects at a host university could lead to the development of a total ship systems center of excellence or a similar initiative that would benefit the community. In addition, the Navy would benefit from the knowledge and research capability available at a university.

A major concern of selecting one university as the host is the potential negative effect on the naval engineering departments at universities not selected. There may be a tendency toward diminished participation in naval engineering research and cooperative participation by the universities not selected as host. There is even the potential for the total loss of naval architecture and marine engineering education and research at one or more of the nonhost universities.

A “neutral” venue, meaning one not located at a university or industry stakeholder, has certain advantages. One is that the administration and contracting functions might be more efficient if done by a professional management services company on the basis of commercial business practices. A second is that the potential negative impact on the universities not selected would be avoided. In addition, there may be better acceptance by and resource support from industry. The shipbuilding industry has had good success with operating a neutral host organization to manage its National Shipbuilding Research Program on shipbuilding process improvement. Depending on the chosen location, a neutral venue might gain better support from the Navy and the professional societies. In fact, the professional resources and technical capabilities of society membership might be best utilized through a research center at a neutral location. One disadvantage of a neutral location would be the difficulty of obtaining the faculty and student resources and support that a host university could provide.

A third choice of location for a cooperative research organization would be a government facility. This location might stimulate support from such Navy groups as the Naval Sea Systems Command and other Navy laboratories. The use of existing infrastructure in a Navy facility might reduce capital investment costs. In addition, the Navy has extensive naval ship engineering research experience and data, which would be available as part of the center’s knowledge base. If the location were a Navy laboratory, the organization might be able to take advantage of its existing focus on ship design,

production, and operational support for the fleet and its concern about the need for well-educated and capable total ship system engineers.

The disadvantage of a government location is that it might be difficult to avoid costly government agency administrative practices and procedures. There also might be a tendency toward less tolerance for taking risks because of the close ties to designing and building ships for today’s fleet. This might be contrasted with pushing technical boundaries, developing new empirical performance data, or developing new materials and production processes. Another disadvantage is that the location might be far from the major university stakeholders, which could create difficulty with close and continuous coordination.

A final factor related to selecting the host location is the perception of how power is being shared among stakeholders. A perception of imbalance can be nullified to some degree by the selection of a neutral location, but such a selection could bring inefficiencies. If a nonneutral host location is selected, the perception can be offset by careful attention to balance in public presentations, websites, and letterheads; by rotating associate directorships; and by rotating the location of annual merit reviews.

For all of the organizational models it reviewed, the committee assumed that ONR would issue one overall funding agreement to the organization, which would, in turn, fund each research project as it was selected. The primary impact of giving this contracting responsibility to a cooperative research organization, rather than having it remain with ONR, is that the administrative responsibility, work, and cost can be transferred from the government to the private sector. The transfer might be either to a university or to a third-party program management group. The committee believes that this will most likely result in an overall improvement in efficiency. It also would place the contract administration function closer to the research work, which would reduce the administrative burden. The contracting resources of ONR could thus be relieved because nongovernment organizations would perform most of the contracting functions.

The federal government has developed a number of contracting vehicles to support cost-shared research with universities and industry. One common vehicle that may be suitable for this initiative is a cooperative research agreement. Such an agreement provides for sharing both the costs and the rights to use the results of the research done under the agreement. This type of agreement has been used by ONR on ship development and technology development programs in the past.

An advantage of using cooperative research organizations and agreements is their ability to accommodate both government and industry funding. Cost sharing between government and industry might improve the relevance

of research themes and results to ship production needs. The use of a single cooperative research agreement between ONR and a centralized research organization might be an efficient way for ONR to contract and administer work because it would transfer administrative burdens from ONR to private organizations. It may also make contracting functions more responsive to the needs of stakeholders.

The day-to-day activities of carrying out a cooperative research program under any of the organizational models requires administrative support, such as procurement, accounting, personnel, travel, computers, and office infrastructure. In all models, these functions would be directed by the host organization. In the case of the professional society model, the society host could provide the administrative support. In the consortium model, the host institution could provide the administrative support. The host institution could be a university, a private firm, a government agency, or a third-party program administrative firm. However, under the project-centered model, because there may not be an existing host organization, it would be necessary to create an organization or contract with a private firm to carry out the administrative support functions.

The three cooperative research models described in this report provide a broad-based organizational structure that can foster naval engineering leadership in both the government and the private sector. The consortium model appears best able to do this, because it can provide a stable organizational structure in which total ship system knowledge and technical management skills can be developed. The professional society model is also broad-based and covers all aspects of ship systems design, but it is more of a loose confederation and relies on significant volunteer effort. The project-centered model also has opportunities for leadership development. However, because it uses separate project groups, it does not provide the broad-based technical and management coordination inherent in the other models. Properly managed, a consortium model would draw in both industry stakeholder talent and the educational skills from university stakeholders.

All of the proposed models encourage leadership by professionals who have an overall understanding of ships and ship systems. Specialists in specific technologies will have an opportunity to grow into systems managers in ship design if they so choose. They can gain broad-based knowledge and a vision of what technical development should be done by understanding the interests of all the stakeholders. With proper motivation, they are more likely

to obtain the commitment of the stakeholders to contribute their best-trained and most capable personnel and resources to the cooperative research.

In summary, visionary program leadership with total ship systems knowledge is not nurtured adequately within the baseline individual investigator model of the current ONR research program. The committee finds that all of the cooperative models, if properly structured and implemented, would encourage the needed program leadership. However, the consortium model appears to offer the most potential for addressing this objective.

Several factors contribute to the ability of all cooperative research organizations to control research quality and maintain it at a high level. First, there are unique problems associated with the small community of technically qualified individuals in the field of naval engineering. This situation makes it difficult to find a group of qualified and conflict-free individuals with balanced biases when seeking reviewers for proposals or a merit review panel. A possible solution is to look outside the immediate specialty area of a research project. This approach has the added advantage of supplementing traditional methodologies with fresh perspectives from other disciplines. Another solution, particularly in the consortium case, might be to encourage the involvement of the practitioner community in the review function. Finally, in the project-centered model, a solution might be to vest quality control authority in the technical review committee.

The panel that reviews research proposals should be carefully balanced so that innovative, high-risk ideas are not screened out and all stakeholders’ interests are considered. Among stakeholder interests are the importance of academic rigor to university researchers and the applicability of the research product to the Navy’s needs.

All cooperative research models include a mechanism to review research effectiveness. An annual review of each research project is considered a minimum for any of the models. The review process in the project-centered model will be the responsibility of the technical review committee. In the other two models, an external review committee appointed or approved by the executive council will meet and review all projects on a regular basis and submit a written report with recommendations for improvements to the executive council. If the review is properly constituted, its value appears to be equal for all models.

In addition to annual merit reviews of projects, postproject reviews of research utilization are important. An executive council should undertake a review of the impact of a completed research project after an appropriate interval. The review should consider the impact on education as well as the utilization of the technology in professional practice.

The leadership of each of the three cooperative research organizations is vested in an executive council, under a variety of names. Alternative strategies in establishing the size, composition, tenure, leadership, and decision-making process of this council will strongly affect the overall success of the organization and the R&D programs it manages. The representation of the three principal stakeholders on the council will affect the degree to which the several constituencies are served and the philosophy, priorities, and direction that the research program will follow.

Too large a council will unnecessarily increase administrative cost and make decision making more difficult. Too small a council will increase the difficulty of adequately representing the diverse views of the stakeholders. If equal representation is desired and the three principal stakeholders are each represented by two persons, the minimum council size would be six. It might be desirable to add other members to the council, such as a professional society representative.

Whatever model is selected, the process and criteria for selecting council members must be carefully considered, along with their tenure and the process for replacing a member who leaves the council for any reason. The willingness to take risks to develop innovative ship design curricula, design concepts, and design techniques should also be a consideration in selecting council members. Knowledge and experience in the following areas should be factors in the selection:

• Early stage total ship design,

• The R&D process and the transfer of technology to ship development programs, and

• The education of naval engineers.

Council leadership is an important issue. Presumably, a council chair will be designated. Whether special powers should be vested in the chair and how the chair will be selected must be decided. The chair could be rotated among the stakeholder communities or might always represent a single stake-holder—academia, the shipbuilders, or the Navy. How the council will make decisions is also an important issue. Unanimity, consensus, or a simple majority might be required. The governing body of the research organization can operate efficiently and effectively if the interests and input of stakeholders are properly considered in the structure that is established. In whatever organizational model is selected, ONR should establish processes to ensure that research funds are fairly allocated and conflicts of interest are excluded from the processes affecting funding decisions, including setting the research agenda. Finally, the selection of dedicated and committed council members is crucial to the success of the enterprise. Those who are asked to serve must

not only represent the range of stakeholder interests, but also be willing and able to devote their time, energy, and talent to making the organization function effectively.

All of the cooperative research models that the committee examined have a positive impact on the education of naval engineers, but their strengths and weaknesses are determined for the most part by how each one is connected to university educational systems and their managers. The ability of the baseline model to improve education depends on the unique interests and departmental influence of the individual investigator. For this reason its overall impact on the educational enterprise is expected to be moderate. The consortium and professional society models would usually establish education committees as components of their organizational structure, whereas the project-centered model probably would not. Because of its project focus, it is expected that the latter model will have a minimal impact on education.

Within the professional society model, an education committee would normally be the channel for research to find its way into university programs and courses. Such a committee is any professional society’s normal access to the academic community. Depending on the nature of the proposed education committee in this model, educational input could take several forms that would require evaluation by a proposal review committee. Within the consortium model, a similar committee would be expected to perform the same functions. The committee finds that this model has a significant ability to affect education in general. It could establish permanent mechanisms and use academic stakeholders in the overall decision-making process. The consortium model has the ability to affect education on a continuing basis.