top of page

Case Studies

Avansys has build a track record for solutions to "no precedent" problems. We create new practice where best practices don't exist. Our engagements range from the application of complexity thinking to define a long term research roadmap for artificial intelligence applications to autonomous flight all the way to reframing Government acquisition practices for rapid development and prototyping. We are most recently advising the product development roadmap for the next virtual conferencing app: Flow OS. We are also advising the development and go-to-market operations for a machine learning-based automatic cost estimating tool. 

The next global virtual conferencing app

[F] OS - the world's fist co-operating platform

We are key advisors to Trencadis Corp as they develop the next virtual video conferencing and collaboration app: Flow OS. We are helping them develop the most complex use case centered around the extraction of context from the information content that builds up in virtual sessions. We are also helping them shape the narrative as part of presentations and marketing pitches to investors. 

 

 

Innovating Government acquisition practices

We guide the FAA on their transformation from all or nothing "fail safe" planning to "safe to fail" incremental decisions. 

Avansys is currently helping the FAA tailor its acquisition practices to the latest product development methodologies and technology delivery models. Acquisition practices heavy on up-front planning overhead have a difficult time keeping up with the pace of commercial technology innovation. In order for the FAA to take advantage of lean start-up, rapid development and other such commercial best practices, it is necessary that systems engineering and investment analysis documentation is reduced. New program controls need to be put in place that acknowledge the shift from all-or nothing "fail safe" planning to "safe to fail" incremental decisions. We are at the forefront of translating the philosophy of agile practices into actionable updates to both acquisition processes and program controls.  

Artificial Intelligence applications for autonomous flight

We developed a clear distinction for NASA between automation and autonomy. In relation to large scale drone operations, we helped NASA focus on autonomy research as key to future drone operations.  

In the 2014 timeframe, NASA felt strong competition from the private sector in terms of a vision and practical approach to integrating drones to the national airspace. The UAS Traffic Management (UTM) working group was established so that government and industry could come together in defining a joint mission that would reconcile the views of drone operators, manufactures and airspace service providers.

 

Avansys supported the definition of the most difficult use case: the scaling of drone operations making use of autonomous flight. We provided NASA Headquarters with a research roadmap for applying artificial intelligence to self-separation applications.  We took inspiration from complexity science's phase transitions to propose a 4-phase evolution roadmap for autonomous drones operations in the national airspace. The work is described in Air Traffic Control Association's Journal article which won the 2017 ATCA article of the year award: A Conceptual Framework for Machine Autonomy.  Our work introduced a clear distinction between automation and autonomy and advised NASA that autonomy is critical for mass drone operations such as might be envisioned by Amazon for home delivery.

​

Say Trevor Tristan, Alessandro Gardi and Roberto Sabatini in their October 2018 "Machine Learning and Cognitive Ergonomics in Air Traffic Management: Recent Developments and
Considerations for Certification" article:

​

No less a body than the Air Traffic Control Association (ATCA) carried Nedelescu’s “A Conceptual Framework for Machine Autonomy” as the lead article of the winter 2016 edition of the Journal of Air Traffic Control [25]. Nedelescu argues that increased automation will only work in a tightly controlled environment; an unpredictable environment—such as one supporting point-to-point drone operations—calls for autonomy rather than high automation. Nedelescu notes that trust in autonomy requires a paradigm shift in VQ&C from a “once-off” design activity to a continuous operational activity—an autonomous machine can develop new behaviors while in operation. He is confident that safety can be achieved in a non-deterministic environment, arguing that one must make allowances for variations in how a machine achieves a valid outcome—particular solutions might contain an element of surprise, the outcome should not [25].

 

Machine learning based automated costing tool

We conceptualized and helped develop a tool for automating cost estimating using state of the art machine learning technology. 

MCR LLC is a mid-sized Government contractor who specializes in business case and investment analysis. 40 years ago MCR pioneered the cost estimating practice as part of the Defense Department push to better project development costs for complex systems. Cost estimating has since become an established discipline with analysts that are certified for various aspects of cost estimation. One of the key cost estimating areas is software. Costing here has traditionally been driven by requirements which are deconstructed into function points which can then be costed using tools such as SEER SIM. Cloud hosting costs are also becoming a significant share of any software solution's development costs.

 

Avansys approached MCR with an idea for the development of a costing tool that uses machine learning to automate what used to be manual tasks performed by a cost analyst. Feeding historical cost data from several software programs, a machine learning algorithm was trained to emulate a manually developed cost estimate in a fraction of the time. What used to take an analyst three weeks can now be done in a matter of seconds by an algorithm. While the result may not be as precise as what a human analyst might produce, the machine learning based estimate provides customers with a very fast rough order of magnitude solution which helps accelerate decisions. The costing tool also includes automated pricing for Cloud services based on the sizing of a hosting solution.

 

The prototype successfully demonstrated that the automated costing tool can be effective in real world applications. The prototype has already been used on several Government programs with outstanding results. Government customers are showing solid interest in using the tool across portfolios of programs, validating the premise for increasing demand for that faster insight into costs, even at a rough order of magnitude level of accuracy. The proliferation of agile development practices within the public sector space is a strong driver for this demand. Avansys is currently supporting MCR with the go-to-market efforts for this innovative new tool. Having shaped the design of the tool and its features, we are now supporting MCR as they build a customer base as it integrates this tool in the company's business model. 

bottom of page