The Zulu Solution

INTEGRATED SLOTS

SCHEDULE EDITOR

BI / REPORTING

PUBLICATIONS & APIs

GATE MANAGEMENT

AI SCENARIO TOOLS

OPS TRACKING

CREW INTEGRATION

Machine Learning AI

Airline planning and scheduling groups frequently have decisions to make that can be difficult to predict, and become quite complex when the implications for an entire schedule are considered.

There are legacy simulation and optimization tools for building a network plan from the micro level of input parameters. However, the goal of schedule timing, aircraft allocation and operational feasibility to achieves optimal performance and profitability is overly complex to be solved solely with explicit programming approaches. Such tools can require a lot of time to set up and run, and are quite dependent on the integrity of the data with as few constraints as possible. The result is often a black box “answer” with little insight into the logic or alternatives for tweaking the solution. If you could know the problem well enough to explicitly program a solution, you would likely already have your answer.

Zulu now offers a machine learning / AI approach to complex problems.  Machine learning was described by Arthur Samuel of Stanford University as “a field of study that gives computers the ability to learn without being explicitly programmed.”  The approach relies on the scientific method of posing hypotheses, providing guidance with “correct” and “wrong” answers and enabling the machine to learn from your data how to predict better and better answers to questions you pose. A key strength of the approach is the ability to use quantitative and non- quantitative data and multiple dimensions that are correlated with the quality of answers. Machine learning's best application is to solve problems that are poorly solved or cannot be solved by numerical means and explicit programming.

Zulu uses a structured approach for working with airline customers to define their strategic question(s), clarify their objectives, frame potential decisions, and model their constraints using their domain knowledge of the operation. Our team customizes the necessary machine learning tools within the Zulu system to yield insights for the decisions to be made. Our developers then integrate the customized framework into the Zulu platform to enable the customer to begin using the system internally as a software-as-a- service for ongoing analysis work following the initial engagement. The Zulu consulting team will facilitate the decision dialog, assist in the adoption of the process and tools, and continue to support ongoing use.

“We were seeking an economical solution from an entrepreneurial team to help us innovate at the rapid pace of our business. They really delivered.”