Imagine a day when a squadron’s flight scheduler arrives at the office and creates the following week’s flight schedule before her coffee even gets cold. Then one of the pilots scheduled for tonight’s inflight refueling exercise calls in sick. She marks the pilot as “unavailable,” and the system recalculates to schedule another pilot who needs night refueling practice. Total time on task: 5 minutes.
It requires immense effort to develop a complex schedule that meets all the training objectives and priorities to prepare a squadron, group, wing, or the Air Force for its primary role. With today’s current tools, scheduling requires one to two people to work all week. Last-minute rescheduling is even more difficult, resulting in scrubbed missions or training. Given that the Air Force is using a myriad of tools and processes to solve this problem with disjointed apps, puck boards, and spreadsheets, complex scheduling tasks are left to be performed in the heads of humans. These are not so much scheduling tools as they are schedule-capture tools; rather than creating a schedule, they only provide a way for the user to capture a schedule already created in her head. In addition, lack of standardization results in a steeper learning curve when aircrews move to new units, making it difficult for schedulers to assist sister units.
How can the Air Force solve this ubiquitous and complex problem?
AI to the Rescue
As part of the AF CyberWorx User Experience (UX) team, we conducted Human-Centered Design (HCD)–focused observational user research at several squadrons and discovered that units typically solve the same scheduling problems with custom solutions, custom apps, custom spreadsheets, and always incongruously. The biggest issue was that regardless of the tools used, the process placed high demands on user cognition, thus relegating each solution to the limitations of each scheduler’s capabilities. A solution to this problem relies on reducing the demands on human cognition by developing a system that can balance the complex needs and priorities to create a workable schedule.
We have identified a single, enterprise-wide Smart Scheduler Paradigm that can be applied to various scheduling domains across the entire Air Force—such as pilot scheduling, Airman training, and readiness forecasting/preparation.
The Smart Scheduler Paradigm (an AI algorithm) puts the emphasis on designing the system to do the work for the human, effectively reducing the cognitive burden of balancing varying assets and personnel needs. After a system is primed with rules-based objectives, assets/resources, personnel, and other adjustable parameters and priorities, the system can automatically generate a suggested schedule. Users will be able to adjust the parameters, such as indicating a pilot is out sick, which can trigger the system to recalculate the schedule using the prioritized objectives. This Smart Scheduler Paradigm will reduce the typical pilot scheduling efforts from about 60 hours per week to about 30 minutes. Other scheduling tasks will likely see similar benefits.
Smart Scheduler Paradigm
The Smart Scheduler task flow is quite simple, only requiring that users define the different elements of the system, objectives, and priorities one time. The system is ready to calculate a new schedule with each parameter adjustment.
Smart Scheduler Task Flow
Poor Technology Maturity
The Air Force’s current approach to technology design and development tends to focus on the technology without much emphasis on the end user. Instead, we adapt the technology to the user. While the rest of the commercial world embraces HCD principles to improve user and organizational successes, the Air Force focuses on technology risk versus the risk of fielding capability that forces the user to adapt to the technology.
Many innovative “solutions” are higher functionality spreadsheets. While many AF members have PhD’s in “Spreadsheet-ery,” users are still left with the cognitive burden of developing schedules while the spreadsheet serves as the knowledge repository. Many spreadsheets are often digital representations of paper or whiteboard tools from the ’60s, ’70s, or ’80s that did little to address the problem or advance the solution. A hallmark of good UX design is getting the system to do more of the work for the user rather than just capturing human outputs.
Commercial companies recognized the cost of non-standard user interaction and interface design models decades ago. They adopt and implement HCD strategies to create cross-platform standardized interaction models that leverage repeatable interaction. These models promote immediate success when a new user tries a new application and reuses familiar interaction models.
Training Addresses the Symptoms, Not the Problems
The Air Force tends to rely on training as a solution for poorly designed technologies. For example, just about every unit completes scheduling tasks with a custom scheduling solution. Every solution is different, requiring new users to waste an inordinate amount of time learning new tools for the same tasks.
It makes sense, then, for the Air Force to commit to developing an enterprise-wide, common scheduling interaction model that could drive the design of every scheduling app. This approach would incorporate greater UX maturity, saving thousands of hours of training time and millions of dollars in development costs.
Institutional Knowledge/Best Practices
A key benefit of a common design paradigm is that it captures best practices from shared institutional knowledge across the Air Force rather than the disparate knowledge specific to individual solutions, driving systemic improvement and efficiency gains rather than incremental improvements at the unit level.
Next Steps … ?
Create a simple mock-up to see how this paradigm can evolve the Air Force into a learning organization and produce a blueprint for other enterprise-wide solution approaches. How many other problems can be solved with such a Human-Centered Design approach?