Reflection and Takeaways on Agility from the SMC-IT 2018 Space Mission Design conference

By Tom Friend – Agile Consultant / LtCol USAF (Ret)

The 6th International Conference on Space Mission Challenges for Information Technology, held in Alcala de Henares, Spain, brought together scientists, engineers, and researchers from NASA, the European Space Agency, universities and industry.

Case studies on how agile methodologies have been applied to mission planning and how scrum has been used in spacecraft construction were discussed, as well as topics such as developing and delivering software, reliability and reuse of software, onboard processing, and communication.

Representing Scrum, Inc. as a keynote speaker, I opened the conference with “Scrum to the Stars” which looked back into aviation history and to the future of innovation in aerospace, and how Scrum methodologies have been, and will continue to be effective tools.

Iterative discovery has been at the core of aviation exploration since the dawn of flight. Whether it was the first aeronauts in balloons, or the Wright brothers at Kitty Hawk, explorers of flight used processes that built on incremental failures and successes. Aerospace design processes were modified as improvements to flight technology were discovered, and the knowledge base expanded. Empiricism and Incremental improvement evolved as a standard path to improvement.  This standard path emerges as patterns.  For example Interfaces in small satellites are deliberately over-designed to reduce need for disruptive renegotiation.  The pattern of S\simple pre-negotiated physical bus structure for data and power increase design versatility, and loose production coupling.  One of the most significant scructural patterns is that of standard adapters allows objects with incompatible interfaces to work together by wrapping its own interface around that of an already existing interface.  These are just some of the patterns that when combined defines the evolving path to improvement.

In essence, Scrum was there at the start of aerospace exploration. Over the years, as systems have increased in size and complexity, common sense has been lost, and projects hit overruns in both time and money spent. By utilizing an Agile framework, you can break down these complex systems into smaller pieces that can then be integrated into the whole design. The step-by-step, incremental approach can be an effective time and cost management tool.

Today the trend in space exploration is making small satellites. Frequently, these small satellites are part of a larger mission.  In doing this, risk is reduced by breaking a complex mission into parts and delivering it in smaller submission components. Think of it as component architecture with your software systems, same pattern. The end deliverable: small satellites that are tailored to a particular mission.

This approach complements Agile planning where focus is on delivering small increments of value and dedicated Scrum teams to build and deliver the satellites.   The success and low cost of small satellites with focused space missions is now mainstream with a standard type of microsatellite called, “CubeSat” that follows set size and weight requirements. This standard is a simple 6-page document keeping with the Agile tradition of minimum viable documentation.

CubeSats by necessity have evolved to leverage many Scrum in Hardware Patterns to speed development and reduce costs. This conformance to patterns has created a whole cottage industry of commercial off the shelf (COTS) suppliers.  They provide hardware and software systems and components that can be used together like LEGOs because they have standard power, size, bussing, and know stable interfaces that allow them to be configured quickly and with low expense.

One of my favorite ways to demonstrate how effective Scrum can be in a hardware setting is a class I give using the CubeSat format. This class is generally offered in a 6-hour format, and is very hands on. In this course, we build a 3D paper CubeSat with a specific mission. All the steps from mission design, roadmap, and components are broken down into a backlog and worked by a scrum team to deliver a fully functional model.  We then walk through the launch and operation of the CubeSat, discussing what each component is doing as it circles the table in the middle of the room that represents Earth.

This simple class exercise using scrum to build components and the visualization of talking through a mission shows how prototyping lets you see problems with design early and builds shared understanding on the team.  These are lessons that you can take back to your own teams to make them even better.

About the Author

Tom Friend is an accomplished Agile consultant, trainer, and coach with 23 years’ experience leading software development teams in various industries to include federal, banking, cable, telecommunications, and energy. He has 12 years of hands on Agile / XP / Scrum software development experience.  He is a distinguished graduate from Air War College and has a BS in Aeronautics.

Project Management and Artificial Intelligence (AI)

By Harold Kerzner, Ph.D.| Senior Executive Director for Project Management, IIL

Recently, I conducted a webinar on PM 2.0/3.0: The Future of Project Management. During the Q&A session that followed, I was asked if PM 4.0 (which I am now researching and will be publishing shortly) would include a discussion of the role of artificial intelligence (AI) applied to project management. I was also recently interviewed by a person working on a graduate degree, who asked what I believed would be the relationship between project management and AI in the future.

It appears that the world of AI is now entering the project management community of practice, and there is significant interest in this topic. While I am certainly not an expert in AI, I became curious about how developments in AI could benefit project management.

A common definition of AI is intelligence exhibited by machines.[1] From a project management perspective, could a machine eventually mimic the cognitive functions associated with the mind of a project manager such as decision-making and problem-solving?

The principles of AI are already being used in speech recognition systems and search engines such as Google Search and Siri. Self-driving cars use AI concepts as do military simulation exercises and content delivery networks. Computers can now defeat most people in strategy games such as chess. It is just a matter of time before we see AI techniques involved in project management.

The overall purpose of AI is to create computers and machinery that can function in an intelligent manner. This requires the use of statistical methods, computational intelligence and optimization techniques. The programming for such AI techniques requires not only an understanding of technology but also an understanding of psychology, linguistics, neuroscience and many other knowledge areas.

The question regarding the use of AI is whether the mind of a project manager can be described so precisely that it can be simulated using the techniques described above. will accomplish this in the near term, but there is hope because of faster computers, the use of cloud computing, and increases in machine learning technology. However, there are some applications of AI that could assist project managers in the near term:

  • The growth in competing constraints rather use of the traditional triple constraints will make it more difficult to perform tradeoff analyses. The use of AI concepts could make life easier for the project manager.


  • We tend to take it for granted that the assumptions and constraints given to us at the onset of the project will remain intact throughout the life-cycle of the project. Today, we know that this is not true and that all assumptions and constraints must be tracked throughout the life-cycle. AI could help us in this area.


  • Executives quite often do not know when to intervene in a project. Many companies today are using crises dashboards. When an Executive looks at the crises dashboard on his/her computer, the display identifies only those projects that may have issues, which metrics are out of the acceptable target range, and perhaps even the degree of criticality. AI practices could identify immediate actions that could be taken and thus shorten response time to out-of-tolerance situations.


  • Management does not know how much additional work can be added to the queue without overburdening the labor force. As such, projects are often added to the queue with little regard for (1) resource availability, (2) skill level of the resources needed, and (3) the level of technology needed. AI practices could allow us to create a portfolio of projects that has the best chance to maximize the business value the firm will receive while considering effective resource management practices.


  • Although some software algorithms already exist, project schedule optimization practices still seem to be a manual activity using trial and error techniques. Effective AI practices could make schedule optimization significantly more effective by considering all of the present and future projects in the company rather than just individual projects.


Project managers are often pressured to make rapid decisions based on intuition rather than by step-by-step deduction used by computers. Nothing is simply true or false because we must make assumptions. Generally speaking, the more information we have available, the fewer the assumptions that must be made. With a sufficient database of information, AI tools could perform reasoning and problem solving based upon possibly incomplete or partial information. AI can visualize the future and provide us with choices that can maximize the value of the decision.

If AI practices are to be beneficial to the project management community of practice, then “pockets” of project management knowledge that existed in the past must be consolidated into a corporate-wide knowledge management system that includes all of the firm’s intellectual property as shown below.


The more information available to the AI tools, the greater the value of the outcome. Therefore, the starting point must be a consolidation of project management intellectual property and the AI tools must have access to this information. PMOs will most likely have this responsibility.


While all of this sounds workable, there are still some downside risks based on which area of knowledge in A Guide to the Project Management Body of Knowledge (PMBOK® Guide) where we apply the AI tools. As an example:


  • As an example, using the Human Resources Knowledge Area, can AI measure and even demonstrate empathy in dealing with people?
  • In the Integration Management Knowledge Area, can AI add in additional assumptions and constraints that were not included in the business case when the project was approved?
  • In the Stakeholder Management Knowledge Area, can the AI tools identify the power and authority relationships of each stakeholder?
  • And with regard to machine ethics, can an AI tool be made to follow or adhere to the Project Management Institute (PMI)® Code of Ethics and Professional Responsibility when making a decision?


While all of this seems challenging and futuristic to some, AI is closer than you think. Amazon, Google, Facebook, IBM, and Microsoft have established a non-profit partnership to formulate best practices on artificial intelligence technologies, advance the public’s understanding, and to serve as a platform for artificial intelligence.[2]


They stated: “This partnership on AI will conduct research, organize discussions, provide thought leadership, consult with relevant third parties, respond to questions from the public and media, and create educational material that advances the understanding of AI technologies including machine perception, learning, and automated reasoning.”[3]


Apple joined other tech companies as a founding member of the Partnership on AI in January 2017. The corporate members will make financial and research contributions to the group while engaging with the scientific community to bring academics on board.[4]


Given the fact that those tech companies are all heavy users of project management, and by some are considered to have world-class project practices, how long do you think it will be before they develop AI practices for their own project management community of practice? The implementation of AI practices to project management may very well be right around the corner.


[1] This definition and part of this blog have been adapted from Wikipedia, The Free Encyclopedia: Artificial Intelligence.
2 (Wikipedia footnote) “Partnership on Artificial Intelligence to Benefit People and Society”. N.p., n.d. 24 October 2016.
[3] Ibid
[4] (Wikipedia footnote) Fiegerman, Seth. “Facebook, Google, Amazon Create Group to Ease AI Concerns”. CNNMoney. n.d. 4 December 2016.

Harold Kerzner, Ph.D. is IIL’s Senior Executive Director for Project Management. He is a globally recognized expert on project management and strategic planning, and the author of many best-selling textbooks, most recently Project Management 2.0.