FROM a management viewpoint, the greatest contrast between the Surveyor and Lunar Orbiter projects was the nature of the relationships of participating organizations, or what might be called the institutional environment. For Surveyor, there was an unusual degree of conflict and friction between Headquarters, JPL, and the prime contractor. For Lunar Orbiter, harmony and teamwork prevailed. Institutions and people worked together in a spirit of mutual respect.
Obviously one cannot generalize from these two experiences on whether harmony or disharmony is more conducive to innovation and the successful management of complex technical projects. What does emerge from the Surveyor and other similar undertakings is that, once engendered, mistrust lingers, coloring the relationships between organizations well after a project has been completed. The manner in which Headquarters, JPL, Hughes, Langley, and Boeing perceive each other still reflects to a considerable degree the impact of the Surveyor and Lunar Orbiter experiences.
The differences in the institutional environment of Surveyor and Lunar Orbiter trace back to the different origins of the two centers and the two prime contractors engaged. Each pair of organizers was characterized by distinctive institutional personalities which influenced relations with their outside worlds.
The Jet Propulsion Laboratory, having entered into its contractual relationship with NASA only a short time before the assignment of Surveyor, was still new to the ways of the space agency. Langley Research Center, on the other hand, was the oldest of the field centers affiliated with NASA's predecessor organization, the National Advisory Committee for Aeronautics. LaRC had been the leading U.S. center for aeronautical research since its formation only 14 years after the Wright brothers' flight. Its senior managers had established close and effective working relationships with their counterparts in nearby Washington, and NASA had given Langley a mission of basic and applied research encompassing the entire range of aerospace programs, both manned and unmanned.
JPL was a leading research and development center in rocketry and missile systems. As it moved into unmanned space exploration, JPL had the difficult task of converting its capabilities to the complex multisystem requirements for space hardware development. The conversion involved major manpower training and redirection.
JPL was accustomed to a high degree of autonomy. Its professional preeminence had bred a strongly independent attitude and a good deal of skepticism concerning more recently formed organizations, including NASA. Its management was quite understandably intent on preserving the scientific and engineering creativity and the independence of its talented staff.
Like the Army Ballistic Missile Agency ( ABMA ) headed by Dr. Wernher von Braun at Huntsville, Ala., JPL was oriented toward the "in-house" approach to development. JPL and ABMA (which became the Marshall Space Flight Center) enjoyed good working relationships with each other, developed largely through their association as the Army team responsible for the first U.S. satellite, Explorer 1, launched in 1958. The two centers had come to share a certain antipathy toward Air Force and Navy approaches to missile development which relied heavily on industrial contracting. Although the high degree of technological innova-tion needed in Surveyor development appealed to JPL's interest in pushing the state of the art, serving as a monitor of an industrial contract weakened JPL's enthusiasm for the program.
Relations between NASA Headquarters and JPL at the senior administration levels were strained from the beginning. In the early 1960's, Administrator Webb and his principal associates, Dr. Hugh L. Dryden and Dr. Robert C. Seamans, Jr., became deeply concerned about failures on Ranger, the JPL in-house project for a hard-landing lunar probe. A congressional inquiry into Ranger, following two high-level NASA reviews, spotlighted some basic weaknesses, including an inadequate system of pre-flight testing, a reflection on past technical judgments. By the spring of 1964, when NASA instituted an intensive review of Surveyor, Headquarters/JPL relations were under severe stress.
The more Headquarters increased its monitoring of JPL projects, the more JPL resented what it regarded as an intrusion on its professional independence. The Headquarters outlook toward JPL was anything but homogeneous because of the differing views at differing levels-the Administrator and his associates and the program offices, particularly the Office of Space Science and Applications and its subdivisions having special responsibility for Surveyor. Headquarters did not present a single institutional front in its relationship with the field, and it was difficult for the field to sort out what seemed quite often to be rather wide divergences. Differences between JPL and Headquarters, moreover, were accentuated as Headquarters began increasingly to concentrate interest and resources on manned flight and the Apollo program. This emphasis was not easily reconciled with the long-standing Caltech/JPL commitment to unmanned space exploration. Such environmental influences operated against good communications and teamwork on the Surveyor project.
A very different situation prevailed in the case of Lunar Orbiter. LaRC is close to Washington, and person-to-person communications between Lunar Orbiter personnel in Headquarters and the field center could be maintained with relative ease. LaRC's managers looked long and hard at the Lunar Orbiter program before they undertook it. They accepted the assignment with full commitment and a determination to make it succeed. The management placed great store in its reputation for fulfilling every mission it set out to accomplish. In reporting to the Headquarters Office of Space Science and Applications, LaRC made no effort to hold back information concerning problems that arose. OSSA reciprocated with full cooperation and support. For all of these reasons, the institutional environment surrounding Lunar Orbiter was favorable to teamwork.
Just as there were marked differences between the Jet Propulsion Laboratory and Langley Research Center, two very distinctive types of corporations served as prime contractors for the respective programs- Hughes Aircraft Co. for Surveyor and the Boeing Company for Lunar Orbiter. The overall experience of Hughes was, in many respects, more relevant to a spacecraft development project than that of Boeing. Hughes' design engineers were recognized for their highly creative talents. But the newly formed Aerospace Group, in which a skeletal Surveyor project staff was located, had limited experience in the management of a complex systems undertaking or in production techniques. For the first several years Hughes Surveyor managers found it difficult to obtain the degree of support and assistance that the project required from other Hughes divisions.
The Boeing Company's organizational approach to the Lunar Orbiter project was quite different. It had accumulated years of experi-ence as a major contractor for production of airplanes and aeronautical equipment. It was familiar with the exacting requirements of systems development. Corporate management was highly sensitive and responsive to requirements imposed by the contractual relationship with a Federal Government agency.
When Surveyor was undertaken, the U.S. space program was still a very young enterprise. Soviet successes in space, beginning with Sputnik I in October 1957, had produced strong pressures in Washington to demonstrate American technological ability to catch up with and surpass the U.S.S.R. The Surveyor program felt the direct impact of these pressures.
It meant many different things to different people. Almost everyone involved, however, saw it as a major program stretching out over a long period and involving several blocks of spacecraft, each for increasingly complex and difficult missions. Managing the program was greatly complicated by the various mutations through which Surveyor passed as it was stripped down to a discrete project dedicated to the support of Apollo.
During the three critical years between the startup of Surveyor and the startup of Lunar Orbiter, the U.S. space program matured and settled down somewhat. NASA Headquarters/field center relationships went through two major reorganizations. The Lunar Orbiter project was the beneficiary of a tremendous effort on the part of NASA Headquarters to develop organizational forms and machinery conducive to effective management. Lunar Orbiter managers could take advantage of what had been learned from Surveyor about techniques and systems of project management. Both NASA Headquarters and the field center applied directly the lessons from Surveyor to the management of Lunar Orbiter.
As an agency, NASA has striven to overcome the temptation to filter the feedback of critical information on past performance. Openness to constructive criticism was espoused. As James E. Webb has observed on the basis of his experience as NASA Administrator, the management of today's large-scale enterprises places a premium on flexibility and adaptation. A continuous and often turbulent process of interaction between a large-scale enterprise and its environment is to be expected, and flexibility in organizational structure is necessary to ride out environ-mental disturbances. Effective adaptation, in turn, depends upon the effec-tiveness of the feedback process.4
Environment is not something apart from, but an integral part of, a project. An effective manager needs to be sensitive and responsive to change in the environment, particularly the kinds of change that alter existing organizational relationships or the relationship between one project and another.
Although the individual manager who moves from one project to another serves as the most efficient carrier of learning experience, NASA fosters the feedback process through manuals and guidelines reflecting past experience. Emerging from both the Surveyor and Lunar Orbiter projects were a number of documents and reports applicable to future project activity. These included NASA publications such as a report on the Surveyor failure reporting system, an article on technology transfer in the Surveyor project by the JPL, project manager, and numerous reports and papers on various aspects of Lunar Orbiter.
At the conclusion of a project, contractor organizations often conduct critiques of their own performance. Although such critiques may contain a high degree of proprietary content, it would be beneficial to the feedback process if such reports, or at least modified versions of them, were made available to NASA upon completion of a project. After the conclusion of Lunar Orbiter, the prime contractor and one of the major subcontractors made such critiques available to NASA, but the Surveyor prime contractor chose not to release to NASA its own internal critical analysis.
ROLE OF INDIVIDUALS
All the principal managers in NASA's Office of Space Science and Applications with responsibility for Lunar Orbiter had also been involved in the management of Surveyor. Perhaps the most significant transfer of learning experience took place among those individuals. At the Jet Propulsion Laboratory, a number of key managers of the Space Flight Operations Facility, and the Deep Space Instrumentation Facility for Lunar Orbiter, were also able to make direct use of what had been learned from Surveyor. Langley Research Center management learned vicariously from the Surveyor experience. The Boeing Co., the prime contractor for the Lunar Orbiter spacecraft, was responsive to suggestions from the Government agency and sought to avoid repetition of mistakes.
In effecting a transfer of learning experience, there is no substitute for an individual manager as a conduit. He carries in his head what he has learned from one experience to another. The individual style and characteristics of managers selected to take on new assignments obviously have a great deal to do with how projects will be conducted. But it is difficult for those administering advanced-technology organizations to determine how managers can best be selected, trained, and rotated.
Management analysts have yet to identify the qualifications that distinguish the ideal candidate for project management assignments from other types of managers. Indeed, a recently completed National Academy of Public Administration study5 found that extensive research and interviews provided no scientific basis for drawing conclusions on the kinds of characteristics, skills, or management styles that best lend themselves to the responsibilities of program or project management. There is even more reason for caution in generalizing on such an issue on the basis of findings in only two undertakings.
The Surveyor and Lunar Orbiter experience might be considered to lend support to the findings of the broader Academy study concerning the difficulty of reconciling different criteria and viewpoints in assessing the qualifications for project management. The Surveyor and Lunar Orbiter findings also support the conclusion of the broader study that individual personal qualities and management capabilities can at times be a determining influence in overall project performance. Most specifically, they conform with that study's emphasis upon "human skills" as the most important of the principal project manager skills. The human skills, which center on the ability to work with others, outranked managerial, con-ceptual, and technical skills.
Human skills and the ability to stimulate effective working relationships between people came much more into play in Lunar Orbiter than Surveyor. The latter, in fact, seems almost to have created an environment of its own which put relations between individuals to the severest test. The pressures and constraints upon Surveyor managers were hardly likely to foster easy cooperation and good working relationships between counterparts. Three-way friction between Headquarters, the field center, and the contractor posed a barrier to good interpersonal relationships.
The impact of the personalities of managers is evident in interrela-tionships with both peers and subordinates. Managers undoubtedly can adopt many different styles to stimulate others to perform. In the Surveyor and Lunar Orbiter experience, assuring that things got done seemed to depend greatly on the power to persuade and a certain ability to "wheel and deal." This was particularly true in the case of the Headquarters program managers. Because they are nominally staff rather than line officials, these managers operate with a somewhat ill-defined authority base. They do not have what NASA calls "directive control" and must confine their role to advisory and monitoring functions while somehow assuring that the program or projects for which they are responsible proceed on target. Although a field center project manager has line responsibility, his real ability to control, like the Headquarters program manager's, is heavily dependent upon his persuasive powers. Those powers need to be brought into play with great skill in the coordination of the activities of other field centers and organizations responsible for various subsystems.
The compatibility of the Headquarters program manager with the field center project managers can be critical to the success of an endeavor. In recent years many of the NASA field centers have come to recognize the importance of this relationship and take it into account in the selection of managers. Headquarters and its field centers now make a joint effort to match the personalities of the two sets of counterparts.
Both Langley Research Center and the Boeing Co. were able to assign managers to Lunar Orbiter who were experienced in prior project activity. The top project managers at Headquarters, the field center, and the prime contractor organization developed smooth-working relationships and highly effective communications with each other. It should be noted again, however, that Lunar Orbiter's discretely defined and technically feasible goals subjected the institutional interfaces of that project to far less strain than was encountered on Surveyor. The smooth-working relationships among various levels of top managers on Lunar Orbiter should probably be regarded as both contributing to and a consequence of successful technical performance. Lunar Orbiter had the advantage of second-generation developments in the three years after the start of the Surveyor program; this also contributed significantly to the high standards set by Lunar Orbiter.
The question of how to achieve good teamwork in project activity involves many intangibles and unquantifiable elements. The difficulty of identifying and measuring the ingredients of teamwork, however, in no way reduces the importance of the concept. Almost all of the Lunar Orbiter managers regarded teamwork as an important aspect of the successful management of that project. In headquarters, the field center, and the prime contractor organization, project personnel regarded their project counterparts with respect and trust. Within both the customer and contractor organizations, moreover, the history of the project was marked by high morale and good teamwork.
Although some sense of teamwork developed in the course of the Surveyor program, it grew slowly and fitfully, spurred by a sense of shared anxiety and concern. The many changes during the project's early years, the basic question whether a launch vehicle would be ready to fly the spacecraft, and concomitant uncertainties about the project's future, were hardly conducive to smooth interinstitutional relations.
The positive attitude and enthusiasm of top management were contagious and infected the Lunar Orbiter project staffs. Some of Langley Research Center's top talents had sought assignment on the project, considering it a career plus. The Lunar Orbiter project organizations at both LaRC and the Boeing Co. were tightly knit cohesive units, yet they operated with full support of and in close communication with functional divisions.
The conditions that prevailed for Surveyor were less favorable. The attitude of most JPL personnel toward a project assignment, particularly one based on contract monitoring, reflected a concern for any diversion from recognized paths of career advancement. There was no doubt about the feasibility of achieving the technical objectives, but the difficulties were tremendous and the Surveyor project was isolated from the main-stream of JPL activity. These factors mitigated against recruitment for the Surveyor project office of some of the best qualified and most talented persons.
The early Hughes organization for Surveyor was highly diffused throughout 13 operating divisions loosely tied to the project office. That office was at a level below many of the divisions on which it was depend-ent, and the Surveyor manager encountered great difficulty in influencing or controlling all project-related personnel. Senior Hughes management was not sufficiently involved in the project to take steps necessary to assure the responsiveness of divisions to project requirements. This was hardly an environment calculated to evoke a strong sense of unity and project commitment. Hughes undertook a major reorganization after Surveyor, to consolidate many activities needing to be under one organizational roof for managing space project activity.
Given the inadequacies in structural formation of the project offices in the Jet Propulsion Laboratory and Hughes Aircraft Co. on top of all the major technical problems besetting the program, it was not surprising that a reciprocal sense of teamwork was slow to develop. Nevertheless, as counterparts worked together, strong ties were forged. For example, the contract manager at JPL and his counterpart at Hughes eventually developed a very effective working relationship. In time, individuals on each side of the fence came to recognize each other's technical competence and skill. With the strengthening of the project organization and the upgrading of management enforced mainly by Headquarters during the latter half of the program, customer/contractor relations improved and a team spirit began to develop.
DEFINITION OF ROLES AND MISSIONS
A good deal of the theory discussed in management literature and a good deal of practical effort to systematize management procedures has been centered on early definition of various roles, missions, and responsi-bilities. Although a period of planning and project definition preceded Surveyor, efforts to carry out the plan ran afoul of many unforeseen con-tingencies. External influences forced the program to go through funda-mental changes in organizational roles and relationships which somewhat vitiated the value of advanced planning.
Quite a few observers have come to believe that a good deal of uncertainty is endemic to research and development activity and that efforts to pin down organizational roles and conform with rigid phasing can be counterproductive. It is argued that too much mechanical effort to build in order and harmony is dysfunctional. In fact, no NASA pro-grams have strictly followed the Phased Project Planning Guidelines issued by Headquarters in 1968. The value of guidelines rests in their utility as points of reference rather than as inflexible standards.
In conforming with the Headquarters policy, both the Jet Propulsion Laboratory and Headquarters seem to have operated on the assumption that the designation of a spacecraft systems contractor implied turning over much of the technical direction of the program to the contractor, and Hughes found that they were not receiving what they regarded as adequate technical guidance from JPL. As the program encountered increasingly serious trouble, Headquarters actively intervened in its management. JPL was compelled to assign a very large monitoring staff to on-site direction of the program. The initially minimal technical direction was replaced by a massive supervisory force. Thus, in the program's latter years, the responsibility for overall spacecraft development was gradually retrieved from Hughes by JPL, thereby altering significantly the respective roles of the field center and the spacecraft systems contractor.
Somewhat parallel changes took place in the management of the Centaur program. General Dynamics Corp., the prime contractor for that program, was originally left very much on its own with a loose monitoring rein. When field center responsibility for the management of the program was assigned to Lewis Research Center in the hopes of pulling Centaur out of serious trouble, Lewis established firm technical control over the contractor; this was a major factor contributing to the successful development of the Centaur vehicle. Another major change in the Centaur program that greatly improved its prospects was the removal of requirements for missions other than Surveyor in the development of the booster capability. The initial decision by NASA Headquarters to assign such an open-ended project as Surveyor to an open-ended launch vehicle made for many complications in both spacecraft and booster development.
Both the Surveyor and Centaur experiences suggest that, during an extended program, roles and responsibilities are not likely to remain fixed or permanent. Arrangements between customer and contractor should be sufficiently flexible to permit each to take advantage of its special strengths and abilities. Adaptive mechanisms to redefine roles and respon-sibilities at various stages of a program are more likely to result in high standards of performance than rigid adherence to a preset pattern.
The Lunar Orbiter experience also demonstrates the positive values of interorganizational flexibility. Informal organizational relationships in the customer/contractor relationships supplemented prescribed formal links. Although Langley Research Center's Lunar Orbiter project office had formal responsibility for "project wide systems integration," the Boeing Co. played an important auxiliary role. With LaRC's tacit approval, Boeing maintained an active monitoring role as a link with the several NASA field centers having a system responsibility in the program.
MAINTAINING ORIGINAL OBJECTIVES
Those who managed Lunar Orbiter at Headquarters, LaRC, and Boeing agreed fully on the importance of adhering to the original objectives. The Surveyor and other space and defense programs offered visible evidence of the risks inherent in changing objectives. The clear lesson was that if you change direction, you will pay for it. The basic objectives of Lunar Orbiter, to obtain data to support the Apollo program for landing men on the lunar surface, remained almost static. As it turned out, the first three Lunar Orbiter missions returned all the data necessary for this set of objectives, and it was possible to add a quest for data sought by the scientific community to the last two flights of Lunar Orbiter spacecraft.
The important consideration from a management viewpoint, how-ever, is that work on the design and development of Lunar Orbiter systems and subsystems was not interrupted by a change in objectives. In the case of Surveyor, the composition of the science experiment payload had been allowed to remain open-ended until late in the program's development stage. In retrospect, it now appears that the uncertainty concerning the number of experiments, their weight and configuration proved to be one of the most serious distractions in the management of that program. Lunar Orbiter managers were careful to avoid this mistake. The Lunar Orbiter Headquarters program manager assured adherence to the principle of minimum change by requiring that his office give prior approval to negotiation of any major change affecting spacecraft design and overall performance.
To reinforce the basic commitment to hold Lunar Orbiter changes to the minimum, management in both the customer and contractor organizations adhered to rigid design review and configuration control programs. After hardware and equipment passed through the critical design review, change was restricted to absolute essentials. Early establishment of a base-line mission for hardware design, worked out between the Boeing Co. and Langley Research Center, greatly facilitated evaluation of the effect of a change. A change board, with representation from each major area involved in a proposed change, reviewed all proposals to assure that only the essentials were authorized. Even before referral to the board, the program manager or the engineering manager had to pass on the submission of the proposed change to the board. These management techniques, together with the basic commitment to make maximum use of "space proven" hardware, made it possible to develop a spacecraft that resembled very closely the design of the original mock-up submitted with the Boeing proposal to NASA.
The major strengthening of organization midway in both the Surveyor and Centaur projects resulted largely from increasing the project staff. In both cases, it had been assumed that the contractor could be given greater systems responsibility than it could exercise. In each case the customer and contractor organizations had started out with small staffs heavily dependent on their respective matrix organizations for technical support. Eventually, more highly "projectized" organizations incorporating all the necessary support functions were developed.
The internal structure of both the customer and contractor organizations for Surveyor went through numerous changes in form and composition. At Hughes Aircraft Co., a major reorganization occurred on the average of every six months. Keeping the interface between structures of the customer and contractor organizations compatible required concerted effort. On both sides, the need for a clear-cut counterpart relationship between key men for every major element of project activity came to be recognized.
At both JPL and Hughes the early Surveyor organizations suffered from the physical dispersion of the activities. Marked improvement came in both project organizations when project personnel were collocated in central facilities at JPL in Pasadena and Hughes in Los Angeles.
The organizational forms used at Langley Research Center and at the Boeing Co. for the Lunar Orbiter program were well suited at all stages to the task at hand. LaRC adhered to its basic philosophy of starting out with a lean organization, essentially as Surveyor began at JPL. But LaRC, unlike JPL, was prepared to supplement the initial project staff, as needed, while also providing full support from other divisions of the center.
The Boeing organization for Lunar Orbiter was highly project oriented from the beginning. Boeing's management had considerable experience in organizing for project activity and was fully prepared to bring together all the manpower necessary for the Lunar Orbiter assign-ment. The tight schedule for the project placed a premium on efficient movement from one phase to the other and for adequate staffing of each phase. Personnel administration provided for timely transition of personnel from design to test and later operational phases. Test and operations teams worked with each spacecraft from final assembly through launch.
The Lunar Orbiter project offices at both Langley Research Center in Hampton, Va., and Boeing, in Seattle, were located from the outset in central facilities where project personnel could work closely together. Close continuing communication both within the two project organizations and between them was a major factor contributing to the success of the program.
There are no firm standards that dictate how far an organization should go in forming project staffs for specific undertakings. The eventual buildup of a very sizable Jet Propulsion Laboratory Surveyor project staff represented a measure that compensated for understaffing in the first half of the program. But the shift is open to the criticism of being an overcompensation, wasteful of scarce manpower. LaRC's organization for Lunar Orbiter, on the other hand, remained lean and relied heavily on the divisional structure. Environmental considerations such as other projects with which participating organizations are involved, the stage of development of an organization, and the availability of the right types of project personnel influence significantly the effectiveness of any form of project organization. The evidence of Surveyor and Lunar Orbiter sug-gests that gradual restructuring and administrative flexibility are necessary to adapt to changing stages. The question of how organizational boxes are arranged, although important and even sometimes determining, is closely tied, of course, to the question of the kind of people who fill the boxes and, particularly, the availability of competent systems managers.
The Hughes experience with Surveyor was one factor leading to a corporate reorganization in 1970. The Surveyor project organization had been located within the Space Systems Division of the Hughes Aerospace Group. Formed in 1961, the division also managed the Syncom communications satellite project. Although the division included many of the technical and managerial elements necessary for managing space projects, it relied on laboratories centered in other divisions of the group for a number of technical requirements. By 1970 the division had established a firm business base. A special predominance in communications satellites had led into other aspects of space communications. The technology and systems management resources were then separated from the Aerospace Group and combined to form a new operational group, Space and Communications, to develop and manage programs in research and applica-tions of space technology. It was designed to comprise virtually all the resources necessary, both technical and business, to conduct these programs.
Systems management capability was scarce when Surveyor was initiated, and few systems managers were available in either the customer or contractor organizations. The real strengths of both the Jet Propulsion Laboratory and the Hughes Space Systems Division resided in creative engineering design talent and researchers in various aerospace specialties.
Systems managers are trained and skilled in supervising the diverse sub-systems of a project in accordance with a schedule to assure integration of the various parts of the project as it moves toward mission fulfillment. A systems manager must have the peripheral vision needed to see the totality of a program, and he cannot afford to focus his attention too long on indepth examination of special areas. He must be able to delegate to specialists in such a way as to assure the highest possible levels of performance in their respective technical areas. Two recognized specialists on systems management, David I. Cleland and William R. King, describe the systems manager as that individual who is appointed to accomplish the task of integrating functional and extraorganizational efforts directed toward the development and acquisition of a specific project. The systems manager is confronted with a unique set of circumstances and forces with each project, and these circumstances and forces channel his thought and behavior into somewhat singular patterns of response.6
Surveyor was a training ground for the development of a sizable number of systems managers highly qualified to apply this skill to future tasks. The structures of the project offices at both JPL and Hughes were significantly altered during the later half of the program to permit more effective execution of the systems management function. This strengthening of the systems function provided better overall integration and represented a major element of the general upgrading of project organization.
When Lunar Orbiter was initiated, more than three years after Surveyor, both Langley Research Center and the Boeing Co. were fully conscious of the importance of systems management. Their project organ-izations included highly qualified systems managers located at the right levels. Boeing was able to assign many of the personnel from two recently concluded projects to Lunar Orbiter, including several highly qualified systems managers; these personnel contributed greatly to the successful management of the program.
In preparing to undertake a complex technical project, a sponsoring agency faces some critical questions concerning the kinds of formal reporting and control systems to apply. How extensive and how detailed should these systems be? How much information is needed at various levels of management?
Everyone recognizes in principle that systematic reporting and control mechanisms are necessary to maintain the discipline required for advanced-technology projects. It is also widely recognized that beyond a certain level, formal reporting systems are wasteful, and that they are counterproductive when they curtail qualified managers' freedom to make decisions. Many good managers insist on being able to make "seat of the pants" judgments without being bound by documents resulting from some formal reporting system.
The Surveyor experience represents an example of an effort begun with too little attention at the outset to the management systems that would be appropriate and the measures necessary to indoctrinate and train personnel in their use. PERT, for example, was introduced several months after the project had started. PERT reporting was handicapped not only by its delayed introduction but also by the fact that the prime contractor, the Jet Propulsion Laboratory, and NASA Headquarters all had had insufficient experience in its use. Hughes Aircraft Co. was not prepared to give up other familiar systems altogether. Although PERT was useful, particularly in the program's early stages, Hughes never fully relied on it for project evaluation and control. Much of the PERT reporting represented more pro forma compliance with NASA requirements than effective utilization of a reporting system for project management.
Surveyor's difficulties stemmed from very fundamental causes such as the changes in the initial program's nature and content and the difficult and complex technical requirements for all major systems, including the launch vehicle. No formal reporting and control systems, however effective, could have overcome the technical difficulties. But, as technical problems were solved and the prospects for meeting all the requirements began to appear reasonable, the management systems required a massive upgrading. Such a substantial overhaul was necessary to assure the degree of rigor and discipline essential to fulfillment of the mission. The up-grading of these systems resulted largely from NASA Headquarters' direct intervention. The Headquarters program manager played a major role in this process.
As a result of intensive and laborious effort, the Surveyor management reporting systems became a true reflection of the state of the project, providing checks in great detail. A trouble and failure reporting system provided not only complete coverage of the technical aspects under review but clear identification of each individual responsible for technical requirements. In the revised reporting systems, heavy emphasis was placed on pinpointing individual responsibility as a stimulus to improving performance. Ultimately, as a result of this type of visibility, a high degree of rigor and discipline was injected into management systems that had previously been too lax and unsystematic.
By the time Lunar Orbiter was started, NASA had made a good deal of progress in refining and standardizing formal reporting and control systems. The issuance of a revised General Management Instruction 4-1-1 in March 1963 clarified the entire field of project organization and management within the space agency.7
Langley Research Center and the Boeing Co. both gave careful initial attention to the adaptation of reporting anod control systems to the project. In contrast to Hughes' resistance to PERT, for example, Boeing accepted the requirement and relied on it as the reporting and control system for all of its work on Lunar Orbiter. Even so, Boeing's Lunar Orbiter program manager made little use of PERT in his decisionmaking. But the system was effective, on the whole, as a device for recording and tracking the status of the project.
Having a great deal of experience in Government contracting that required extensive formal reporting and control, Boeing management sought from the beginning of Lunar Orbiter to keep the volume of reporting from becoming excessive and the reported information from being unnecessarily redundant. Yet NASA reporting requirements for the project exceeded what Boeing considered the optimum level of detail. Midway in the project, Boeing was able to convince LaRC that some of the reporting requirements could be discontinued, thereby reducing the cost.
Both LaRC and Boeing took care to assure that management report-ing systems were updated and well maintained. Boeing sought to make the reports true and meaningful indicators of the state of the program. By keeping the reporting systems in good repair, those responsible for Lunar Orbiter were able to avoid the need for a massive upgrading. Lunar Orbiter managers made information systems come close to serving the basic purpose for which they were intended-to communicate the essential information on the state of a project to all those who needed to know in both the customer's and the contractor's organizations.
What stands out in the Lunar Orbiter experience, however, is not the overriding importance of formal reporting but the optimal use of informal person-to-person communications. Lunar Orbiter experience corroborates the conclusion reached in Richard Chapman's study: "No formal arrangement can replace the dynamic system of personal and informal relations developed by key members of the project team to meet that project's particular needs."8 The compatibility of individual managers serving the customer and the contractor helped greatly to assure Lunar Orbiter's success.
ROLE OF HEADQUARTERS
Only rarely does a program require the extent and depth of inter-vention by Headquarters that occurred in the case of Surveyor. Both the Surveyor spacecraft and the Centaur on which it depended faced such serious troubles that the highest levels of Headquarters management felt compelled to intervene. The story of Centaur demonstrates the importance of decisionmaking at the Headquarters level.
Marshall Space Flight Center was the first NASA field center to be assigned responsibility for Centaur after the transfer of the program in 1959 from the Air Force. Many factors worked against the interests of Centaur at MSFC. Senior management at MSFC focused its attention mainly on the development of the powerful Saturn launch vehicle for Apollo, and the demanding responsibilities for Saturn left somewhat limited technical and managerial resources available for Centaur.
Before being assigned to MSFC, Centaur had gone through numerous changes and shifts in objectives, and there were numerous technical conflicts in the propulsion requirements represented by several different potential customers for a single launch vehicle. Advent, a military communications satellite project, imposed demands on Centaur that were incompatible with the Surveyor requirements, and a year elapsed before the Advent mission was deleted.
Even after Centaur was transferred to MSFC, the Air Force retained responsibility for monitoring the prime contractor. In the face of many serious technical difficulties associated with Centaur development, MSFC's top management concluded that it would not be feasible to meet the minimum weight-lifting requirements of Surveyor and that the program should be cancelled in favor of a Saturn C-1/Agena combination. The Jet Propulsion Laboratory concurred in the MSFC recommendation.
Headquarters, after carefully reviewing the situation, confirmed its position that the Centaur concept was both technically feasible and essential to the launch vehicle program for the space effort. It thus rejected the recommendation of senior management at MSFC and JPL. Responsibility for Centaur was transferred abruptly then to Lewis Research Center. This was interpreted as a rebuke to MSFC and a signal to the other centers that they could not back out of major development commitments assigned by Headquarters.
On numerous occasions Headquarters felt compelled to intervene in Surveyor. For example, a major Headquarters investigation of the program Op. cit in early 1964 uncovered many serious weaknesses in both technical and managerial aspects of the project and led to a series of correctional moves. The Headquarters review contained detailed proposals for tightening and upgrading project organization and management both at JPL and at Hughes. Headquarters urged JPL to appoint a Deputy Director who could help in JPL administration and management while keeping an eye on Surveyor. A former general manager of the Atomic Energy Commission was designated by JPL to serve in a similar capacity. He instituted significant changes in the business administration and management practices of JPL in general and Surveyor in particular.
This appointment was highly charged with internal political over-tones. Headquarters senior administrators were dissatisfied with the general management at JPL and saw the difficulties encountered in Surveyor as an opportunity to force a change. JPL senior administrators, on the other hand, were skeptical of any organizational or personnel changes inspired by NASA. Thus, despite the new deputy's substantial contributions to improved management, he left JPL and accepted a position outside NASA before the seven Surveyor flights were completed.
In several instances, difficulties within the Surveyor and Centaur project organizations became so serious that representatives of general management were designated to assume direct dayto-day responsibility for management. A representative of JPL's senior management served as Surveyor project manager for a critical period: the Deputy Associate Administrator in Headquarters Office of Space Science and Applications acted in the capacity of Surveyor program manager; and, for several months, the Director of the Lewis Research Center was project manager of Centaur. Ideally, a project, once assigned to a responsible field center, would not require such penetrating intervention by Headquarters. Only a monitoring function was needed on Lunar Orbiter.
In view of the eventual success of Surveyor and more than a dozen other projects that were simultaneously sponsored in the area of space science and applications, the overall record is impressive. Cost escalations, however, were not uncommon and many projects slipped behind schedules.
By intervening in Surveyor, Headquarters helped reduce constraints for which it shared a considerable degree of responsibility. The original underestimation of the complexity of the Surveyor program, the imposition of manpower and financial ceilings, prolonged insistence on an unreasonably open-ended combination of scientific experiments for the payload, the many changes in scope and objectives of the program, and the tying of Surveyor to an unproven launch vehicle were all problem-causing factors that were attributable to decisions made by Headquarters. Only Headquarters could effectively ameliorate them.
The Jet Propulsion Laboratory, as the responsible management center, was slow to accord the Surveyor project the priority that Headquarters wanted it to receive. The deep concern of top JPL management caused by the series of troubles encountered in the Ranger project and the requirements for other in-house projects limited JPL's efforts on Surveyor. It took a major Headquarters review and persistent Headquarters directives, both orally and in writing, to bring JPL management to improve the Surveyor project organization.
The Headquarters review in the spring of 1964 also pinpointed a number of deficiencies in the Hughes Aircraft Co. organization. Headquarters instituted a direct watch over Hughes operations to assure that more support was being given to the project and that more attention was being given from senior levels of Hughes management. Headquarters continued to be dissatisfied with aspects of Hughes management and technical performance well into the operational phase of the project.
When all the demanding tasks involved in the Surveyor lunar landing missions were complete, the aftermath was characterized by institutional friction. In each of the three principal organizations involved in the project-NASA Headquarters, the Jet Propulsion Laboratory, and Hughes-the Surveyor personnel tended to view their own organization's contribution as the critical key to success. For the record, each organization has formally acknowledged that team effort was essential to ultimate success. But among Surveyor personnel in the three participating organizations there is far less willingness to acknowledge the contribution of other groups and individuals than there is among the participants in the Lunar Orbiter project.
The Surveyor spacecraft systems contract was awarded on the basis of a source evaluation by JPL, and JPL negotiated the contract with Hughes. The contract was written as the cost-plus-fixed-fee (CPFF) type, and was converted to an incentive basis quite late in the program-on the day before the launch of the first Surveyor spacecraft. JPL's administration of the CPFF contract failed to keep pace with the many change orders and modifications, and fell far behind in its accounting of the financial status of the project. About a year of intensive work in the Surveyor contract office was needed to upgrade contract records. At about the same time, JPL, in response to Headquarters direction, began efforts to persuade Hughes to convert to an incentive contract. Although Hughes at first resisted, strong Headquarters insistence induced Hughes management to accept the new contract. When the project was completed, the company earned fees totaling several million dollars more than their minimal expectations under the CPFF contract.
Both customer and contractor management then regarded conversion of the contract as a highly beneficial administrative measure well worth the massive effort entailed. The entire work breakdown structure and financial reporting system had to be revised as part of the total conversion process. After the conversion, however, it was possible for the first time in several years for customer and contractor to operate on the basis of mutual agreement on the status of the contract.
More important, the incentives had a highly beneficial impact on Hughes' performance. The prospect of earning fees tied to specified and realistic cost and schedule targets motivated all levels of personnel. General management at Hughes had played an active role in negotiating the contract conversion and took steps to assure that the entire project received full support. An award fee in the new contract provided additional incentives for high standards of performance in the management and operation of the project. This fee stimulated maximum effort in all areas of project management over and above those that had a direct relationship to costs and schedules.
The original negotiated cost of the Hughes contract for seven space-craft was $67 million. Final Hughes contract costs came to $365 million, over a fivefold increase. For Lunar Orbiter, the original negotiated cost of the spacecraft contract was $84 million and the estimated final contract costs at $144 million represented less than a twofold increase. There is no incontrovertible method of correlating the cost performance on Lunar Orbiter with the fact that it was the first major NASA flight program to be undertaken on the basis of an incentive contract. Although the incentive fees were generally regarded as a positive feature in the contractual relationship between customer and contractor, the Boeing Co. representatives attribute less significance to incentives than to the strong corporate determination to achieve success in their first spaceflight project.
The Boeing contract for Lunar Orbiter was cost plus incentive fee, whereas the two major subcontracts with Eastman Kodak and RCA were CPFF. Boeing's management had anticipated that, once having negotiated the prime contract, they would be able to persuade the two subcontractor firms of the advantages of an incentive form of contract. Both Eastman and RCA held out firmly against what they considered an untested and risky method of contracting. The absence of incentives in the two major subcontracts tended to undercut the impact of the incentives in the overall spacecraft system development.
As the first major NASA project to be awarded on an incentive basis, Lunar Orbiter broke important new ground in the development of standards for determining and administering fee awards. Both Lunar Orbiter and Surveyor experiences attest to the positive value of incentives. NASA's early favorable experience with incentive contracting on such projects as Lunar Orbiter led the agency to increase the use of this type of contract to the point where it represented 68 percent of total award obligations for external research and development in 1968.
More recent years have brought a shift away from incentives, down to a level of 46 percent in 1970. This shift reflects growing awareness that dollar profits may be less of a motivating force in a private organization's performance than the impetus to hold a place in a growth market or the need to assure corporate survival. Despite this decline in the relative importance of incentives, Lunar Orbiter contracting experience was worth while because it helped to inform NASA about effective approaches to research and development procurement.
Analysis and interpretation of cost data relating to space projects is a complex task involving many variables. Assuring complete objectivity is difficult. From the viewpoint of achieving the goals of the national space program, what matters is essentially the ratio of costs to the amount of scientific and space engineering information produced in each project. Did the spacecraft send back the kind of data that it had been designed to retrieve? Were the data useful to the scientific community and to engineers and technicians engaged in other ongoing space activities? Measuring by these criteria almost inevitably involves subjective judg-ments concerning the utility of the data returned.
NASA's original estimated total cost of the Surveyor project was $125 million whereas the final costs came to $469 million, somewhat less than a fourfold increase. Lunar Orbiter costs were first estimated at $77 million and wound up at $163 million, or slightly more than a twofold increase. To gage these two records of cost performance, it is useful to compare them with other NASA projects in unmanned space exploration. An of 16 research and development projects being conducted by OSSA during the sixties indicates that the average final costs were somewhat less than three and a half times the initial estimate.9 Extended delays in several of the early OSSA projects as well as increases in the number of spacecraft flown contributed to substantial cost escalations in several of the earlier projects.
Costs, of course, are in considerable part a function of time. The nearly fourfold increase in the total cost of Surveyor over the original estimate reflects the fact that the project took more than two years longer than originally estimated. Previous discussion has brought out many factors contributing to delays in Surveyor. The early planning for Surveyor was highly unrealistic and vastly underestimated the complexity of the task.
Lunar Orbiter final costs, in contrast, were only slightly more than double the original schedule, the first launch being made within two months of the initial target date. The preceding discloses how Lunar Orbiter managers took advantage of the three years of learning experience that elapsed between the start of the Surveyor program and the initiation of their project. The Lunar Orbiter record compares favorably with the overall OSSA performance.
Both Surveyor and Lunar Orbiter were highly successful from the viewpoint of gathering scientific and engineering data essential to the Apollo program and future lunar exploration. The data on the chemical composition, density, and bearing strength of the lunar surface acquired by means of Surveyor's instruments were essential to the planning of Apollo landings. The photographic data acquired from both Surveyor and Lunar Orbiter formed the essential basis for selection of initial Apollo landing sites. The acquisition of these data and their systematic exploitation called for an effective relationship between the hundreds of consulting scientists, engineers, and technicians engaged in the projects. The manner in which this relationship evolved is discussed in the following section.
Meshing the interests of scientists and engineers in the Surveyor program was a real challenge and the source of much management difficulty.
Communications barriers between scientists and engineers reflect the differing motivations and orientations of the two disciplines. The scientist tends by and large to be interested in acquiring knowledge about his special field. For him, the mechanical means for attaining that knowledge may be of only incidental interest. The engineer or technician, on the other hand, is likely to be primarily interested in the mechanics of an instrument problem. In a gross sense, he focuses his interests on the "how to do" rather than the "what to do." Though he wants and needs to know enough about the scientific objectives of an experiment to do his engineering work satisfactorily, he is essentially concerned with the very prac-tical issues of what will work.
There are also likely to be wide divergences among scientists seeking data from a spacecraft. One scientist does not necessarily have much interest in the work of other scientists whose experiments may be riding on the same space "bus" as his. Only in the later phases of the Surveyor program were the principal investigators and other science advisors for Surveyor brought into full realization of the interdependence of the various experiments
NASA Headquarters, where the selections of scientific experiments for Surveyor were made, was pressed by the scientific community to pursue many different lines of investigation via Surveyor. These pressures made Headquarters reluctant to narrow the options for change. Indeed, in the early years when Surveyor was conceived as a three-block program, there was good reason to plan for a broad and diverse science program. But it now appears to have been quite unreasonable for Headquarters to have insisted that the design of the spacecraft be such as to accommodate any combination of some 30 science experiments, particularly when most of the experiments were also continually being changed.
To avoid some science/engineering problems, the Jet Propulsion Laboratory kept the Surveyor science investigator teams somewhat removed from the technicians and engineers responsible for instrument design. JPL was concerned that scientists might disrupt the work of the engineers and that some engineers might become overly committed to perfecting a scientist's pet experiment. In time, the need for such concern diminished. But in the early years of Surveyor, the slight concern of the scientific investigators for the impact of their experiments on spacecraft performance caused trouble for project managers.
Whenever science is an important aspect of an engineering task, the scientific objectives must be clearly recognized at the outset. Special management attention should be given to those levels of the engineering organization at which the science inputs are made to be sure that they are properly incorporated in the payload. The scientists responsible for the experiments must work closely with the engineers responsible for the basic assembly on which the experiments will ride.
The photographs and other data on the lunar surface returned by Surveyor aroused great interest in the scientific community. NASA Headquarters and the Jet Propulsion Laboratory developed highly effec-tive machinery for collecting, analyzing, and widely disseminating scientific data. Thus the Surveyor data served the interests of both those responsible for planning the Apollo landings and the growing ranks of scientists interested in information on the Moon.
Lunar Orbiter had no major science objectives until they were added for the last two flights. The U.S. Geological Survey was the only outside group involved, and Langley Research Center needed only a small science complement. The Lunar Orbiter consequently had fewer problems than Surveyor and less need for the elaborate organizational structure that was established for Surveyor, a project in which more than 100 outside scientists and a highly sophisticated science division at the responsible field center were involved. Nevertheless, the photographs returned from Lunar Orbiter and the data collected on the last two flights provided a rich store of information that is still being widely studied and analyzed by lunar and other scientists.
4 James E. Webb, Space Age Management: The Large Scale Approach. McKinsey Foundation Lecture Series, sponsored by the Graduate School of Business, Columbia Univ. (McCraw-Hill, 1969), pp. 149146.
5 Richard L. Chapman, with the assistance of Robert H. Pontious and Lewis B. Barnes Project and Program Management in NASA: The System and the Men, National Academy of Public Administration (Washington, 1971), pp. 165-168.
6 David I. Cleland and William R. King, Systems and Project Management (McCraw-Hill 1981), p. 12.
7 NASA Management Manual, General Management Instructions Number 4-1-1, subject: Planning and implementation of NASA Projects (Mar. 8, 1963). This instruction replaced GMI 4-1-1 of Jan. 18, 1961.
8 Op. cit.
9 Final cost estimates for 16 OSSA projects totaled approximately $2.5 billion in comparison with original estimates of about $784 million. (From Memorandum from Assistant Administrator for Program Plans and Analysis to the Administrator, subject: NASA Project Cost Projections (Apr. 10, 1969).