Acta Astronautica ] (]]]]) ]]] ]]]

Contents lists available at SciVerse ScienceDirect

Acta Astronautica
journal homepage: www.elsevier.com/locate/actaastro

Improved cost monitoring and control through the Earned Value Management System
Howard Hunter n, Richard Fitzgerald, Dewey Barlow
The Johns Hopkins University Applied Physics Laboratory, 11100 Johns Hopkins Road, Laurel, MD 20723, United States

a r t i c l e i n f o
Article history: Received 22 December 2011 Received in revised form 9 August 2012 Accepted 18 September 2012 Keywords: Earned value Project management Cost Benet

abstract
As economic pressure and competition for budget among federal agencies has increased, there has been an increasing need for more granular data and robust management information systems. This is especially true for the execution of major civilian space programs. This need has resulted in new program management requirements being implemented in an attempt to limit cost and schedule growth. In particular, NASA Procedural Requirements (NPR) 7120.5D requires the implementation of an Earned Value Management System (EVMS) compliant with the requirements of American National Standards Institute (ANSI)/Electronic Industries Alliance Standard 748-B. The Radiation Belt Storm Probes (RBSP) program management team at The Johns Hopkins University Applied Physics Laboratory (JHU/APL) made a decision to implement an EVMS on RBSP during Phase Ba year earlier than specied in the contractual Phase C reporting requirement as dened in the NPR. This decision was made so that the project would have the benet of 12 months of training and hands-on implementation during Phase B. Although there were a number of technical and process hurdles encountered during Phase B and into Phase C, the system was working well when the Integrated Baseline Review (IBR) was held in August 2009. The IBR was a success because it met the review requirements. It was also clear to all IBR participants that the EVMS was providing value to the project management team. Although the IBR pointed out some areas of concern regarding process and ANSI compliance, the system had markedly improved the projects ability to monitor cost and schedule. This, in turn, allowed the project team to foresee problems in advance, formulate corrective actions, and implement course corrections without causing signicant adverse impact to the project. Opponents of EVMS systems often communicate the unfavorable opinion that EVMS systems create unnecessary cost and administration. Although it is undeniable that EVMS implementation does not occur without cost, the cost is minimal in comparison to the benets of successful implementation. This paper will focus on the implementation of EVMS on the RBSP project, explain EV processes and the implementations cost, and analyze the benets of EVMS to provide insight into cost/benet considerations for other projects considering EVMS implementation. This paper will do this by focusing on the following points: (1) RBSP is the rst full-up implementation of earned value management (EVM) at JHU/APL; (2) RBSP EVM started in Phase B; (3) RBSP EVM implementation has been working well in Phase C/D; (4) RBSP EVM implementation has been recognized by Goddard Space Flight Center and NASA Headquarters as successful;

Abbreviations: ANSI, American National Standards Institute; CAM, Control account manager; EV, Earned value; EVM, Earned value management; EVMS, Earned Value Management System; IBR, Integrated Baseline Review; IMS, Integrated Master Schedule; JHU/APL, The Johns Hopkins University Applied Physics Laboratory; NPR, NASA Procedural Requirements; PM, Program manager; PMB, Performance Measurement Baseline; RBSP, Radiation Belt Storm Probes; SD, Space Department; SMD, Science Mission Directorate; VAR, Variance Analysis Report n Corresponding author. Tel.: 1 443 778 2965; fax: 1 443 778 6515. E-mail addresses: howard.hunter@jhuapl.edu (H. Hunter), richard.tzgerald@jhuapl.edu (R. Fitzgerald), dewey.barlow@jhuapl.edu (D. Barlow). 0094-5765/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actaastro.2012.09.010

Please cite this article as: H. Hunter, et al., Improved cost monitoring and control through the Earned Value Management System, Acta Astronautica (2012), http://dx.doi.org/10.1016/j.actaastro.2012.09.010

H. Hunter et al. / Acta Astronautica ] (]]]]) ]]] ]]]

and (5) an assessment of the benets of EVMS to the project management team and sponsor shows that the systems benets outweigh the cost of implementation. & 2012 Elsevier Ltd. All rights reserved.

1. Intent of the Johns Hopkins University Applied Physics Laboratory (JHU/APL) Earned Value Management System (EVMS) The goal of the JHU/APL EVMS is to provide Space Department (SD) management with a consistent, standard framework for assessing project performance. The basic tenets of earned value management (EVM) have been introduced and used within the JHU/APL SD on the Radiation Belt Storm Probes (RBSP) project. The EVMS currently interfaces with the systems that comprise the accounting system and scheduling tool and incorporates data from both to produce the output necessary for earned value (EV) data production. These data are comprised of an Integrated Master Schedule (IMS), Contract Performance Reports, a Variance Analysis Report (VAR), and monthly schedule reports. This output has allowed the SD project management team to make timely decisions regarding the project, provided a sound basis for project cost estimates and funding requirements, and assisted in meeting the project reporting requirements for the contract cost/schedule performance measurement data. 2. Beginning the implementation Before the start of the implementation, there was concern stated by the control account managers (CAMs) who were going to use the EVMS. Our CAMs were concerned about the potential impact of the requirements that would be levied on them. The principal CAM concerns included the possibility of the EVMS taking up too much of their time or the system not supplying any added value to them. With that said, there were factors that precipitated an EVM implementation at JHU/APL. In the period from 1979 to 1996, JHU/APL was singularly successful in arriving at program costs that were within a few percentage points of planned costs. Eight spacecraft bus programs were within 7 8% of the initial cost estimate at Phase C/D start. Recently, though, missions and instruments have experienced cost growth. Along with this, requirements were levied by the sponsor [NASA Procedural Requirements (NPR) 7120.5D] regarding cost and schedule requirements in an EVMS environment. To combat future cost/schedule growth and meet the reporting requirements, JHU/APL began implementing several new processes and systems, one of which is the current EVMS. The EVMS implementation started to become a reality after several key factors came together. JHU/APL management took a series of steps to support the implementation. Senior management promoted a cultural shiftamong senior department managers through project managers to the CAMswith the use of an EVMS to manage a project. This buy-in from senior management expedited the change process.

The implementation team used this momentum and built on it by demonstrating to users that there was value in the EVMS. The EVM team used a gradual approach for the implementation by rst using the EVMS on smaller departmental projects. This approach demonstrated to the project manager (PM) and CAMs that the EVMS was benecial in the management of the project. It also allowed the EVM team to leverage the project to bring on different aspects of the EVMS and rene them. Other factors in this approach included maintaining open communication among the project team and senior management, introducing different aspects of the system, and allowing the users time to digest how to use the system and to garner CAM support. This approach was critical for instituting EVMS on RBSP. The lessons learned from the smaller projects were used and additional key factors were identied to keep the implementation on track. One factor was showing value to the user. The EVMS team demonstrated the value of an IMS to the CAMs. This was done by working with the CAMs through workshops. These workshops built on the relationship between the EVMS team and CAMs to break down the cultural barriers of change through the analysis of the IMS and its data products. This helped ensure that the CAMs would have a better understanding of the IMS and its products. It allowed the CAMs to identify key variances, what causal effect was behind each variance, and what could possibly be done to mitigate the variance. The EVMS used a one system approach. This meant the entire project team at JHU/APL collaborated on the efforts of the implementation with the sponsor and instrument teams. The EVMS at JHU/APL used MS Project le updates and cost data from partnering institutions as input. The project did not force individual EVM systems on each partnering institution. Our partners were mostly universities that did not have this capability. By using this approach, we were able to maintain a one system approach. The EVMS team fostered open communication via continuous discussions with the CAMs, project team, and sponsor organizations regarding the implementation and what it meant to them. Frequent feedback to the team on data quality, results, and areas for improvement facilitated continual improvement of the system throughout implementation. Having using this graduated approach and demonstrating initial success, the implementation was poised to move onto its next phase: the use of EVMS in a reporting phase on a SD mission. 3. EVMS in Phase B Key components of the EVMS were developed in Phase B on RBSP. The baseline was developed for both schedule and cost. An Integrated Master Plan and IMS were put in

Please cite this article as: H. Hunter, et al., Improved cost monitoring and control through the Earned Value Management System, Acta Astronautica (2012), http://dx.doi.org/10.1016/j.actaastro.2012.09.010

H. Hunter et al. / Acta Astronautica ] (]]]]) ]]] ]]]

place. Work was authorized via Work Statement Authorizations. Data began to be reviewed on a monthly basis with the CAMs in a workshop-type format. Reviews were held with the JHU/APL and sponsor project teams and eventually culminated in a project baseline review. This review was intended as a precursor of a scheduled Integrated Baseline Review (IBR) to be held during Phase C of the project lifecycle. This baseline event was two and a half days long. Disruption to the project was kept at a minimum. Review requirements were met with successful reviews of the IMS and Performance Measurement Baseline (PMB) (cost) and the identication of signicant technical, schedule, and cost risks. The sponsor summarized the success of this review by stating good progress had been made on EV processes and that cooperation and support is strong between JHU/APL and the sponsor. The other keys to success were tailoring the EVMS to the needs of the CAMs, leveraging existing project management processes, and maintaining a collaborative environment. Tailoring to the needs of the CAMs helped generate support from this group. This support was a key factor in the success of the EVMS in this phase. The CAMs learned to trust the quality of the data and appreciated the minimal overhead required to obtain them. CAM support was obtained while meeting the challenge of staying within budget for the implementation and use of the EVMS.

 Providing a solid foundation for mutual understanding of project risks.

 Providing the project management teams with a

thorough understanding of the project plan and its risks and allowing for early intervention and the application of resources to address project challenges. Increasing condence in the project PMB. Continuing and developing the EVMS further through Phase C/D of the project.

 

As noted earlier in the paper, the IBR was a success. The sponsor noted that the EVMS had become a tool for good project management, that continuous progress was being made, and that the IMS had greatly improved from the baseline review held the previous summer. Here was concrete evidence that an EVMS was delivering on the promise of increased project control and a better understanding of where project risk was. How were these data being received at an agency level? Let us nd out. 5. EVMS, CAMs, and the sponsor The RBSP project has entered Phase D of the project lifecycle. The spacecraft has come together with major subsystems in place. Integration and test has started for the instruments that will y on the spacecraft to provide mission science, and the EVMS continues to measure project performance and report on risk at the subsystem level. So, has the EVMs delivered on the intent statement noted previously and has there been real benet to the project? Let us take a look. Initial statements from the CAMs expressed a sentiment that the EVMS may not provide benet to them or potentially distract from their technical functions. By working closely with the CAMs regarding schedule development, holding frequent interactive meetings with them, fostering mentoring relationships with them, and making them part of the EV team, this sentiment has changed. Comments have ranged from, the EVMS has allowed insight into trending across the project and enhanced the management decision process, to the EVMS has helped to pinpoint areas where development was behind schedule and there was potential risk for cost growth, to the benet of oversight far outweighs the burden of implementation. This sentiment has been echoed at the agency level for NASA. On a number of separate occasions, the agency has passed along praise of the JHU/APL EVMS. Within the Science Mission Directorate (SMD), comments such as not only does the JHU/APL EVMS generate the data for the project, more importantly the project team uses the data to manage its project were backed up by the fact that the SMD gets independent assessments from the Independent Program Assessment Ofce and Aerospace verifying these statements. It is good news indeed that our CAMs, PM, and sponsor have reached concurrence that the EVMS has added value to the project management of a major NASA mission. But the question remains at what expense.

4. EVMS in Phase C and an IBR Building on the implementation and success generated in Phase B, the RBSP project moved into the full reporting and project build cycle of Phase C. This was done keeping several tenets in mind, including that the agreement reached between stakeholders (functional supervisors and CAMs) regarding the use of resources would be honored, best practices and processes would be developed and matured, EVMS would not become a time burden to the CAMs, EVMS would not become a punitive tool, analysis of project data would be managed by exception by paying attention to the variance drivers (done through an improved ability to drill down to areas driving the variance), and the information from the EVMS would be timely and readily available. The goal in this phase was to have the EVMS become part of the tool belt of the CAMs and project team in assessing project performance. In taking these steps, it was hoped that the CAMs and project team would use more of the EVMS capabilities and discover the value added of the system. The expectation was the EVMS would provide early detection of problems, enabling faster response and corrective actions to emergent issues. As Phase C progressed, an IBR was scheduled by the Living with a Star (LWS) Program Ofce. There are many goals that need to be met in order to have a successful IBR, but a couple of the keys for what success looked like were as follows:

 The IBR process would benet both the NASA LWS

Program Ofce and the JHU/APL RBSP project team by

Please cite this article as: H. Hunter, et al., Improved cost monitoring and control through the Earned Value Management System, Acta Astronautica (2012), http://dx.doi.org/10.1016/j.actaastro.2012.09.010

H. Hunter et al. / Acta Astronautica ] (]]]]) ]]] ]]]

6. Cost and benets of the JHU/APL EVMS There have been studies on the marginal cost of maintaining the EVMS criteria on a major procurement. Costs range in these studies from initial implementation efforts to the cost of maintaining an EVMS within the American National Standards Institute guidelines for a compliant system. Within the studies, costs can range up to 5% of the contract cost [1]. Other studies have looked at what drives the cost of an implementation to meet the intent of the guidelines. Some of the cost drivers pointed out included excessive documentation, excessive levels of detail in the work breakdown structure, and written variance analysis reports [1]. The estimated cost at completion, for the EVMS, on RBSP is 1.22% of the total estimated project cost. This estimate is well within the range of previous studies conducted on the costs of an EVM process. The question remainswas the implementation and the cost of having an EVMS on the project worth it? The answer to that is yes. The key elements for success of the EVMS on RBSP had to do with the project teams ability in planning, monitoring, and controlling the project with the right balance between process and results. Stakeholders on RBSP realized that this balance would not be easy to achieve. What made the EVMS work were the steps taken to achieve this

balance. The EVMS team never lost sight of trying to show value at every step in the process. This was done by showing value to the CAMs by having quarterly Estimate to Complete reviews, VARs written by the EV team and edited by CAMs/PM, the EV data available in a server environment for ease of access, and monthly schedule data input with EV personnel ready to interface with the CAMs and to provide feedback. Demonstrating this added value to the CAMs and to the project management team fostered an environment that was effective in controlling against cost and schedule issues that arose. This level of control has allowed the project to stay within its Contract Target Cost throughout Phase C/D of the project lifecycle. The insight EVMS gave into issues and the way it gave stakeholders the ability to be able to clearly understand what was behind a potential issue made for more effective management decisions. The EVMS became a management tool, not just an accounting tool. It became one more tool in the tool belt of an effective project manager/CAM. Perhaps that is the true benet of an EVMS and a measurement of its success. Reference
[1] D.S. Christensen, The costs and benets of the Earned Value Management Process, Acquis. Rev. Q. 5 (Fall) (1998) 373386.

Please cite this article as: H. Hunter, et al., Improved cost monitoring and control through the Earned Value Management System, Acta Astronautica (2012), http://dx.doi.org/10.1016/j.actaastro.2012.09.010