automating  airport snow removal 

Snow removal is a tedious, dull task that amazing crews take on in the winter months. Inspired by the bad weather in 2015, my team decided to focus on automating snow removal, and to start at the airport level. We were particularly interested by Logan Airport (and implementing the system at other Class B airports) as larger airports tend to have a bigger budget. However Logan is a pretty efficient system with A/B crews on 6 hour shifts. One of the interesting scenarios we looked at was how to automate snow removal at smaller airports (which had more temporary hires and usually dealt with the same crew potentially on the field for 12+ hours, depending on the length of the storm). 

Getting a robot to follow a path is pretty straightforwards. Doing that reliably in all weather is hard. Getting people to trust autonomous vehicles is another problem entirely.

systems  analysis

This project consisted of a variety of analyses such as:
  • Stakeholders/Needs Analysis
  • System Definition/Concept of Operations 
  • Technology Roadmap
  • Gap Analysis
  • Proposed Solution
  • Risks of Development
  • Failure Modes and Effects

Purpose

To perform a systems analysis of opportunities for integrating autonomy into existing airport snow removal systems.

Concept  of  operations

The System's mission is to: 
  • Improve Safety
  • Reduce Operating Costs
  • Be as effective as the current system
Modes of Operation
  • Training
  • Repair/Maintenance
  • Recovery
  • Normal Snow Removal Operation
Proposed Autonomous Vehicle Scenarios
  • Crewed Semi-Autonomous Vehicles
  • Leader-Follower System
  • Crewed Fully-Autonomous Vehicles
  • Fully Autonomous Vehicles

Key  findings

  • Winter Storms in 2015 caused over 7,300 commercial flights to be canceled, which cost an estimated $230M to the US economy
  • Equipment is costly
    • Multipurpose machines ~ $1M
    • Snow blowers ~ $500K
    • Sand spreader + plow ~ $250K
  • Winter preparation lasts year round, and vehicles are kept operational for 8 months out of the year
  • FAA recommends airports such as Boston Logan to remove 1 inch of snow in 30 mins.

Gap Analysis

  • Tech gap: Technology is advancing rapidly but is years from actual implementation
  • Artificial Intelligence gap: Challenges with quantifying human decision making processes make reliable algorithm development challenging
  • Human Factors/Interaction Gap: Uncertainty in what makes the ideal interface or what interface may work for this application

Summary

Associated costs with delays and equipment
  • Training
  • Operations
  • Maintenance 
  • Flights (and any delays)
Existing Stakeholder needs
  • Improve runway safety
  • Reduce worker fatigue
  • Reduce operating cost
  • Maintain snow removal efficiency/effectiveness

some of our charts

Random other facts learned:
  • JetBlue is the largest carrier at Logan with anywhere from 30-35 planes stationed overnight. Their hangar is only big enough to fit 3 planes at any given time, therefore their mechanics are outside in all weather 365 days out of the year from midnight-4 AM. Compelling case to look into for our next systems analysis!