We performed eleven interviews over the course of the spring semester. Finding appropriate CIs proved difficult, because our focus dealt with planning in a very specific environment: people from different backgrounds working to create a plan based on constantly changing information. Furthermore, missions such as Phoenix are no longer running so it was not possible to conduct a CI in this environment. For this reason, interviews proved to be extremely valuable in accessing the information we needed. By speaking with people from many different roles on a mission, we were able to gather different perspectives and form a detailed understanding of the organization and the interactions between mission planners, engineers, and scientists. The majority of our interviews investigated the Phoenix and Mars Exploration Rover missions. We also conducted a number of interviews in similar areas, both inside and outside of NASA, in order to understand how planning is done in other domains.
In order to observe first hand the challenges involved in coordinating between different people, and planning for actual rover traverses, we conducted six contextual inquiries both inside and outside of NASA. The Mars Desert Research Station (MDRS) gave us insight into interactions between experts in different domains, and the Mars Exploration Rover mission allowed us to see how a current, long term, rover mission is run.
We conducted four contextual inquiries at the Mars Desert Research Station to learn more about how a group of diverse specialized experts, who have not previously worked together, behave and operate on a simulated mission. In this simulation they lived together and could not interact with the outside environment except under special circumstances. The team members came from different academic disciplines, and had different science goals on the simulation. We sought to witness how a diverse group of scientists worked together to execute plans, as a valuable analog to interdisciplinary group collaboration on NASA missions.
Each individual on this station had a designated role as well as their own experiments. However, almost all experts on the station needed some other involvement by members in order to accomplish tasks. Although participants had specific roles, the roles were not adhered to.
We conducted two contextual inquiries with personnel on the Mars Exploration Rover mission. We sought to learn how a real life distributed Mars rover mission operates on a daily basis. This was done by observing panoramic camera (pan-cam) payload uplink leads (PULs) on the MER mission. We contrasted this information with that received while interviewing MER mission planners. Also, we were interested in seeing how a mature mission operates compared to other shorter duration missions.
The pan-cam PUL’s role is to create rover sequences for the pan-cam that get uploaded to the daily rover from high level plans. In order to accomplish this, instrument sequencers have to start with a high level plan established that morning given to them by the Keeper of the Plan (KOP) for the day, refine it, discuss the plan with other sequencers, and create the sequences. This is done throughout the day with constant open teleconferences to keep in touch with all the individuals involved in the mission. We observed the instrument sequencer for the whole process, except for some meetings that we were unable to attend due to security reasons.
A broad survey into collaboration literature helped provide several insights into areas that affect collaborative work, including cultural differences, establishing trust given distributed collaboration, and collaborating through digital medium. An in-depth look into planning for surgical suites provided an analogous planning domain with similar resource constraints. Operating rooms must coordinate hospital resources, patient health, and surgeon availability in order to reduce downtime.
The research revealed both problems with collaboration, and potential solutions for those problems. The research made explicit possible communication breakdowns. This was helpful in completing interviews and contextual inquiries because the research prepared us for noting sources of communication breakdowns, exploring plans in alternative domains, and in comparing the process of maximizing scarce resources to processes observed in the Mars mission-planning environment.
We compared twenty popular commercial collaboration tools to see what features were commonly available and to provide us with an overview of what collaboration issues currently exist and how they are addressed. In order to understand the purpose of each piece of software, we answered four questions about each: Who is using this tool? What are they using it for? What are the main features of this tool? What collaboration issues does this tool address? Viewing the tools’ functionalities revealed design opportunities, because many important aspects of collaboration identified in our research are under-employed in popular business software.