What. Build forecast models for future product demand. Quantify interactions of market players. Measure price elasticity, design and interpret A/B tests.

How. Econometric techniques and expert market insights. Differentiated product and logit demand specifications.

What. Apply state-of the art machine learning techniques for business transaction outcome predictions.

How. Probabilistic neural networks with task-tailored architectures, decision trees and other methods. Technology stack such as XGBoost and Tensorflow Probability.

What. Simulate business processes at atomic level. Optimize business decisions against simulations. Build custom dynamic decision engines for real-time deployment.

How. Simulation stack inspired by reinforcement learning applications. Optimizations with deterministic algorithms and randomized solution space search methods such as genetic alogirithms and reinforcement learning.

What. Design and implement data pipelines to feed reports and dashboards with actionable analytics. Rapidly deploy data lakes for project purposes.

How. Technology stack such as Amazon Web Services, Apache Spark and Kedro.

What. Forecast revenue, cost, cash flow and valuation metrics for capacity enhancement projects.

How. Use of volume drivers from own demand models. Extrapolation of historic unit performance and benchmarking to industry peers.

What. Forecast traffic for purposes from short-term hourly operational dispatching to long-term master planning. Build future design day schedules and assess capacity bottlenecks in the sub-systems.
‍
How. Field-tested set of forecasting and modeling techniques in end-to-end Python data pipelines. Use of industry guidelines such as International Air Transport Association Airport Development Reference Manual.

What. Plan passenger and cargo network and fleet for aircraft ordering and future business sizing.

How. Simulations and numeric optimizations as end-to-end Python data pipelines. Differentiated product modeling of market share and demand on new routes.