As discussed in Chapters 2 and 3, each PI of a PI-led mission must propose a management organization and plan within the constraints set out in the AO. That plan is reviewed as part of the selection process and adjusted during Phase A to ensure compliance with NASA standards and requirements. No two management plans are alike because there is such a variety of team structures within these mission lines. This chapter describes the roles that NASA plays in PI-led mission management activities at both the program and the project levels. PI management roles and approaches are also discussed.
PI-led mission teams respond to an AO that solicits creative management plans. At the same time, NASA specifies many requirements on programmatic reporting and schedule, cost, and technical implementation that have developed over the long history of space missions. NASA documented these requirements in NPR 7120.5B, revised recently as NASA Procedural Requirements (NPR) 7120.5C: NASA Program and Project Management Processes and Requirements. All PI-led missions use the latest version of the NPR document as either a guide or requirement when responding to the description of project management responsibilities contained in the AOs. NPR 7120.5C describes the general responsibilities for project implementation, including technical and programmatic reviews—for example, peer reviews, a mission-level PDR, a mission-level critical design review (CDR), a pre-environmental review (PER), a pre-ship review (PSR), and launch readiness reviews (LRRs) and the general reporting requirements between project and program offices.1 The individual programs and projects have the responsibility of integrating the details of these implementation and reporting requirements into their projects. (See Chapter 3 for more details of the life cycle of a project.)
During the proposal stages, it is the PI’s responsibility to establish the various management roles and responsibilities for each organizational member of the team and the individual members. These roles and
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences 4 Management of PI-Led Missions As discussed in Chapters 2 and 3, each PI of a PI-led mission must propose a management organization and plan within the constraints set out in the AO. That plan is reviewed as part of the selection process and adjusted during Phase A to ensure compliance with NASA standards and requirements. No two management plans are alike because there is such a variety of team structures within these mission lines. This chapter describes the roles that NASA plays in PI-led mission management activities at both the program and the project levels. PI management roles and approaches are also discussed. BACKGROUND PI-led mission teams respond to an AO that solicits creative management plans. At the same time, NASA specifies many requirements on programmatic reporting and schedule, cost, and technical implementation that have developed over the long history of space missions. NASA documented these requirements in NPR 7120.5B, revised recently as NASA Procedural Requirements (NPR) 7120.5C: NASA Program and Project Management Processes and Requirements. All PI-led missions use the latest version of the NPR document as either a guide or requirement when responding to the description of project management responsibilities contained in the AOs. NPR 7120.5C describes the general responsibilities for project implementation, including technical and programmatic reviews—for example, peer reviews, a mission-level PDR, a mission-level critical design review (CDR), a pre-environmental review (PER), a pre-ship review (PSR), and launch readiness reviews (LRRs) and the general reporting requirements between project and program offices.1 The individual programs and projects have the responsibility of integrating the details of these implementation and reporting requirements into their projects. (See Chapter 3 for more details of the life cycle of a project.) During the proposal stages, it is the PI’s responsibility to establish the various management roles and responsibilities for each organizational member of the team and the individual members. These roles and 1 Pre-environment review takes place before the hardware is put in a thermal vacuum test environment; a pre-ship review, before the entire spacecraft is shipped to the launch site.
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences responsibilities are typically documented in the final proposal on which selection in based. Each program office then works out the details with each PI and his or her team soon after selection to ensure that all the roles are filled and responsibilities assigned. From then on, the individual projects take on a life of their own, developing along unique paths with their individual casts of players. NASA PROGRAM OFFICE ROLE The program offices of the PI-led mission lines participate in managing their projects. At the time the committee gathered data, all but the Explorer Program Office appeared to be in a state of flux or definition. In part because of the Explorer Program’s long history, the roles and responsibilities of program management are most clearly defined in that program. Explorer Program Office The Explorer Program Office is both located at and relies on NASA GSFC. It has about 38 full-time equivalent (FTE) employees to manage and support three MIDEX missions, three SMEX missions, and three MoOs. GSFC also carries out major system reviews for the projects, such as the PDRs and CDRs, through its Office of System Safety and Mission Assurance. The Program Management Council (PMC) at GSFC evaluates each Explorer project’s readiness to proceed into implementation (see Table 4.1), including independent cost estimates through its research analysis office, and makes its recommendation to Headquarters. The Explorer Program Office participates in its missions’ development by assigning to each mission a civil servant mission manager, who is the point of contact with the program, serves as the contract technical officer, arranges for requested NASA support, and keeps NASA Headquarters familiar with the project. The Program Office has system engineering, financial, and scheduling/planning experts who support the mission manager. The mission manager also has a separate budget outside the project cost cap that he or she TABLE 4.1 Independent Evaluations Related to PI-Led Missions Review Who Description Independent reviews NASA, independent, and NASA-approved PI nominees Independent reviews called by PMCs at NASA Headquarters or designated NASA centers to monitor the development and implementation of PI-led and other missions via gateway reviews (e.g., PDR, CDR) and other more-focused technical assessment reviews as needed. Independent Technical Authority (ITA) Warrant holders in technical areas (NASA employees) ITA is a NASA-wide independent process rather than a review for providing authority, accountability, and responsibility for technical requirements, processes, and policy at NASA. Warrant holders provide program/project managers with technical requirements and decisions and resolve technical issues. The application of ITA to PI-led missions has not yet been fully established. Independent verification and validation (IV&V) NASA IV&V Facility, West Virginia Project software is assessed to locate risks in the development and operation of software for NASA missions and projects. These assessments can include systems assessments, which focus on basic requirements, design, testing, and system processes in development; life-cycle assessments, which identify risks and recommend actions for any software development life cycle aimed at ensuring mission success; startup assessments, which locate the software elements of higher risk and complexity pertaining to safety and recommend IV&V. The IV&V Facility reports to GSFC. Peer reviews Internal and external experts Working-level, independent engineering reviews that examine subsystems in detail and address problems in advance of system-level reviews. SOURCE: Available at <ivv.nasa.gov/forprac/asses/assessment.php>.
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences can use to bring in technical support from the broader GSFC and NASA community.2 Lines of authority among the PI, the PM, and the mission manager have to be regularly reinforced, communicated, and respected throughout the partnership at all times. The PI must be included in every major decision affecting the mission, where what is “major” must be determined by communication. In an effort to make all parties aware of the lines of authority for Explorer missions, an “insight agreement” is written during Phase B and added to the contract between the PI and program. This document defines the specific roles and responsibilities of the mission manager in providing support to both the project as well as the program office and includes any reporting and interfacing he or she requires from the PI and PM and their team to maintain insight into the project. To avoid making the interface with the mission manager an added burden on the project, it is necessary to create and maintain a real partnership. It is important for proposers of Explorers to be well aware of practices such as insight agreements as they formulate their mission plan. All projects have different levels of complexity and difficulty, and project teams vary in their experience and capabilities. The Explorer Program Office independently assesses the capabilities of the project teams and tailors its management and involvement in each project accordingly. Examples of mission manager contributions include scheduling/planning support, ground system development, oversight of local instrument providers, and interfacing with the GSFC test facilities. In fulfilling these roles, the mission manager is expected to report to the PI or the PI’s PM. The mission manager also reports to the manager of the Explorer Program, who shares the fiduciary responsibility to see that each project stays within the PI’s committed cost and schedule. The program manager supports the mission manager and project, but if problems arise will call for a special review or recommend a termination review3 (see Chapter 7) to Headquarters if (s)he feels it is warranted. The Explorer Program Office also reports monthly to the GSFC PMC on all of its projects. In this way it is both a project supporter and an independent overseer discharging NASA’s fiduciary responsibility. Conflict is possible when traditionally competing institutions must work as a team. The Explorer Program Office has been able to work effectively with institutions competing for the same missions as GSFC PIs and teams, although sensitive information can make this a challenge. A conscious effort must be made to fence individual projects within GSFC. After selection of a project, any conflicting institutional management styles must be resolved within the project, but that is necessary regardless of the institutions involved. In the case of the Explorer Program, the GSFC style of management naturally prevails, but it is generally accepted at an agreed-on level by PIs and PMs as part of the organization of this mission line. Discovery Program Office The Discovery Program Office (DPO) was originally located at NASA Headquarters, with primary oversight being supplied by the program executive and/or the program scientist, supported by Headquarters personnel. This arrangement changed in the late 1990s, when Headquarters moved many activities and responsibilities to NASA centers. The DPO was set up at the NASA management office co-located at JPL, with limited support being provided by non-civil-servant JPL employees. This setup limited the functionality of the DPO, which needed a noncompeting entity to review nongovernmental organizations’ proprietary information. For much of the late 1990s and early 2000s, the DPO was staffed with approximately 1.5 to 2 FTE civil servants to oversee all Discovery projects.4 In 2004, it was moved to MSFC as the Discovery/ 2 Of course this approach must be exercised with much sensitivity with regard to control and decision making in the PI-led projects. 3 A mission can be subjected to a termination review if it exceeds the cost cap, excluding cost growth that is beyond the PI’s control. 4 NASA hired the Aerospace Corporation in place of JPL to support the Discovery Program Office for a period of time.
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences New Frontiers Program Office. The size and functionality of the Discovery/New Frontiers Program Office at MSFC were still being defined as this study was under way. The major difference between the DPO and the Explorer Program Office concerns the role of a NASA center and its level of involvement in a project’s management. The support system that a NASA center can provide is significant. In the Explorer Program, GSFC has been used (as reported by PIs and PMs interviewed by the committee) in support of the projects, especially by project teams that are not able to carry out all of the functions needed to implement a space mission. In the Discovery Program, the emphasis has been on selecting a PI team that already has the ability to play all the roles for conducting a space mission. As a result, Discovery projects have not needed as much program support as Explorer projects other than what they get from certain government facilities—for example, launch vehicles and the Deep Space Network. NASA oversight was originally envisioned to be minimal and, for many in the scientific community, this vision helps to maintain a PI team’s control over its own project. The differences in the roles of the program offices in the missions are reflected in the AOs for the Explorer and Discovery programs. The AOs for Discovery contained language that implied less NASA oversight and more freedom for the PI to independently manage his/her project than was implied by the language contained in the AOs for Explorer missions (see Appendix C). While the DPO, like the Explorer Program Office, functions as the project advocate within Headquarters, it has had considerably less influence than the Explorer Program Office, which has had the backing of GSFC. The DPO still provides the same essential functions as the Explorer Program Office—interfacing with other NASA and government agencies for facilities support; arranging independent review committees (in conjunction with Headquarters) for programmatic, technical, quality, and safety issues; and tracking project schedules and budgets. However, until now, the DPO has provided these services with far fewer staff and less support from its center, a situation that may change with the recent relocation of the Discovery/New Frontiers Program Office to MSFC. As NASA oversight in general increased in the late 1990s and early 2000s in response to a series of mission failures and mishaps, much of the independence originally enjoyed by the Discovery missions started to erode. This erosion was noted (and seen as a negative change) by many PIs and PMs for Explorer, Discovery, and New Frontiers projects. NASA Headquarters’ oversight in the form of independent reviews (see Table 4.1) appeared to increase the most for those within the Discovery projects, probably because their program office was less structured. The DPO, with its increasing number of projects, also was viewed as overburdened. Many Explorer project PIs and PMs interviewed by the committee said that they had taken advantage of the services provided by the Explorer Program Office. In comparison, Discovery PIs and PMs indicated that staffing and infrastructure issues reduced the level of support provided by the DPO for their projects. Mars Scout and New Frontiers Program Offices The Mars Scout Program Office is located at JPL and the New Frontiers Program Office was recently combined with the Discovery Program Office and assigned to MSFC, as described in Chapter 2. Since these are relatively new mission lines, the committee has no comments on their approaches or organizations. PI TEAM MANAGEMENT ROLES Based on its interviews, it was clear to the committee that many different combinations of mission team leaders and distributions of responsibility can be made to work for the PI-led mission mode. However, a common theme was the importance of a good relationship between the PI and the PM, and the experience base of the PM in particular. Chapter 2 describes the PI and PM roles in PI-led missions. Two management
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences approaches, one for Explorer (as embodied in the Far Ultraviolet Spectroscopic Explorer) and one for Discovery (in the Comet Nucleus Tour mission), are presented below. These two approaches illustrate the unique circumstances faced in the management of all PI-led missions. PMs and PIs must be prepared to deal with many diverse technical, administrative, political, economic, and/or personnel-related challenges. The documentation and sharing of lessons learned from the cases described below relate to overall aspects of PI-led mission management as well as the specific items confronted. Far Ultraviolet Spectroscopic Explorer The Far Ultraviolet Spectroscopic Explorer (FUSE) mission was the first Explorer mission to be developed in the current (true PI-led) mode (see Chapter 2). However, FUSE differed from later MIDEX and SMEX missions in several ways. First, its development costs were somewhat higher, perhaps by a factor of 1.5, than those of MIDEX missions. Likewise, the technical and organizational complexity of the mission was higher than that of MIDEX missions. FUSE is a spectroscopic observatory with subarcsecond pointing designed to serve the broad astronomical community. During the prime mission phase, slightly more than half of the observing time went to guest investigator projects selected by NASA, with most of the remaining time assigned to the PI science team. During the subsequent extended mission phase, which began in April 2003, all of the observatory time has been made available to guest investigators. The instrument consists of four telescopes and spectrographs but was built as a single unit. Thus, the common descope option of deleting an instrument was not an option on FUSE. Both Canada and France contributed flight hardware before the constraints and rules of International Traffic in Arms Regulations (ITAR) went into effect, producing a new environment. The satellite was integrated in a series of steps, starting with the construction of the detectors at the University of California, Berkeley; the integration and checkout of the spectrograph at the University of Colorado; the integration of the instrument and then the satellite at the Johns Hopkins University Applied Physics Laboratory (APL); and, finally, environmental testing at GSFC. During Phase A and most of Phase B, the PI was responsible for only the instrument and the PI science program (in the spirit of the early facility-class Explorers mentioned in Chapter 2). The decision to develop the mission in the current PI-led mode was made late in Phase B as part of a substantial restructuring of the program to reduce cost and schedule. From that point on, the mission was implemented in the same way as other PI-led missions, with a cost cap and a 35-month schedule for Phases C and D starting in November 1996. Despite these differences between it and other PI-led Explorer missions, the FUSE mission illustrates one way a PI-led mission can be carried out from a university campus. NASA assigned responsibility for all aspects of the FUSE mission, including the instrument, spacecraft, and operations, to the PI, with GSFC taking responsibility for oversight, but not management, of the development process. In this case, the PI delegated technical, schedule, and budgetary responsibility to the PM. The PI was the lead scientist with responsibility for all scientific aspects of the mission, except for the guest investigator program, which was the responsibility of a NASA project scientist assigned to the original mission. The PI remained deeply involved in all aspects of the mission. Although the Center for Astrophysical Sciences (CAS) of the Johns Hopkins University had developed instruments for space missions previously, it had not undertaken a space program of the size and complexity of FUSE. To develop the mission, CAS made extensive teaming arrangements with APL, several commercial firms, and other universities. It took responsibility for mission management, mission architecture, systems engineering, the telescope optics, science team management, and the satellite control center. In practice, many of these positions were held by team members at Johns Hopkins. The availability of experienced engineers and technical managers for dealing with issues was an important factor in the
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences success of the development process. The heavy use of Internet documentation, peer reviews, a standing senior review board, and other generally recognized good practices were followed. Many of the teaming organizations provided personnel with prior experience with FUSE subsystems for the integration and testing. Owing to technical problems, primarily in the spacecraft attitude control system, mission launch was delayed to June 1999. The development cost was about $118 million, $10 million more than the planned cost at the beginning of Phases C and D. Although designed for 3 years of operation, the mission is now in its sixth year. As of early 2005, 51 million seconds of data had been obtained for more than 525 scientific programs, yielding over 3,600 spectra for more than 2,300 different objects. There are about 300 refereed publications based on FUSE data in the scientific literature, and they continue to appear at a rate of about six per month. In addition, more than 40 articles based on the design and development of FUSE instruments have appeared in the scientific literature. FUSE technology was the basis for the design of the Cosmic Origins Spectrograph, presently awaiting deployment. Overall, the PI-led management system for FUSE can be said to have succeeded, although it must be noted that the greater investment in development than experienced by later MIDEX projects could have been a factor in that success. Comet Nucleus Tour Mission The Comet Nucleus Tour (CONTOUR) mission was the sixth mission in the Discovery Program. It was less costly than other Discovery missions, with the exception of Lunar Prospector. CONTOUR was fairly typical in its management structure. The PI was from Cornell University, which had extensive experience in providing spaceflight instruments but none in providing spacecraft or in managing space missions. The PI teamed with APL to provide the spacecraft, mission design, flight operations, and overall project management. APL provided an experienced PM, who was approved by the PI. The PI also brought other NASA centers onto the project team (JPL for navigation expertise and GSFC for spaceflight instrumentation), and other commercial and academic team members were contracted and managed from APL. Cornell University had the responsibility for education and public outreach, the science team, and science operations (including data analysis). In accord with the AO, NASA assigned responsibility for all aspects of the mission to the PI. During the course of the CONTOUR project, the DPO moved from NASA Headquarters to JPL. Both the PI and PM interfaced with NASA primarily through the DPO, but there was also a significant interface with the program executive and the program scientist at NASA Headquarters. NASA funded the project via a grant to Cornell University and a contract with APL, both administered by GSFC. Other government funding of institutions (e.g., JPL, GSFC) was handled through internal transfers in NASA executed by the DPO, as approved by the PI. The PI provided significant oversight of the management of the mission but delegated day-to-day activities to the PM at APL. There were frequent communications not only between the PI and the PM but also between the PI and other critical team members at APL, including an APL mission manager, the mission system engineers, and the optical system engineers. The PI had worked extensively with these individuals on other projects prior to proposing CONTOUR. The PI also routinely communicated with upper management at APL during the course of the project. The project conducted an extensive review and test program, as approved and accepted by NASA. In coordination with the Discovery Program manager, NASA independent assessment team reviews were held in parallel with the normal project reviews that had been scheduled prior to the new review requirements (e.g., PDR, CDR). Extensive peer reviews, many with outside reviewers, were also carried out. Around the midpoint of the program, software independent verification and validation (IV&V) were added to comply with the new NASA requirements. These new requirements were imposed first and later negotiated for their
OCR for page 31
Principal-Investigator-Led Missions in the Space Sciences cost impacts. Because of the launch constraints of this deep space mission, schedule impacts were not allowed. CONTOUR had some early costing problems, but these were resolved before the confirmation review. After confirmation, the schedule and budget remained on track, with budget adjustments for changes of scope over the cost cap (for new requirements). Unfortunately, the spacecraft was lost several weeks after launch on August 15, 2002 (see Chapter 6). Changes in Management and Oversight NASA is an evolving organization that learns from past experience, events, and new information and responds to administrative interests and factors (e.g., ITAR, the Independent Technical Authority (ITA), and internal initiatives). Particularly in the late 1990s and early 2000s, NASA made many changes to its oversight philosophy and its views on the acceptability of risk in response to a series of failures and mishaps (not, however, in PI-led missions). Lessons learned and recommendations from committees5 that studied these problems were incorporated into new agency-wide requirements that affected all NASA missions, including PI-led missions. The major changes that NASA mandated and that were incorporated into subsequent AOs include the addition of software IV&V and new risk management and cost reserve requirements. These were to be incorporated into all active projects, regardless of their development stage. Many of these changes were imposed in the middle of the implementation phase of some PI-led missions, which had very little flexibility to incorporate such changes given the constraints on their budgets and, often, their launch schedules. The PI project teams worked with their respective program offices to deal with these impacts, but in general these externally imposed changes meant significant disruptions for the PI-led missions. In addition to the new requirements cited above, NASA program offices and/or Headquarters added oversight in the form of formal reviews beyond those agreed upon after selection (see Table 4.1). As aversion to risk exposure increased within NASA, projects reported that they were increasingly required to present formal reviews on various technical or management issues raised by NASA officials. As the committee learned during its data-gathering process, the impact of these formal reviews on the project team—the burden of preparing, holding, and following up on action items—was significant in the view of many PIs and PMs. They said these formal reviews (as opposed to peer reviews) had minimal value and were burdensome because they took time away from critical project activities. NASA oversight requirements continue to change. During the committee’s period of analysis, NASA announced a new oversight group, the ITA. Because ITA was created as a NASA response to the Columbia shuttle disaster, safety and technical reliability are its two main areas of concern. ITA’s plan is to use a limited number of NASA-center-affiliated designated experts, called warrant holders, to function as a clearinghouse for specific safety and reliability requirements. Its impact on the tightly cost- and schedule-constrained PI-led missions is, as yet, unknown. 5 For example, the NASA Integrated Action Team (NIAT) reviewed NASA findings on the failures of the Mars Climate Orbiter and the Mars Polar Lander, integrated the lessons learned, and augmented them with findings from its own examination of project management practices across the agency. For science missions, the NIAT report led to large increases in the amount of documentation required, additional independent reviews, mandated use of IV&V software, and an intense focus on risk analysis and management (NASA Integrated Action Team, 2000, Enhancing Mission Success: A Framework for the Future).