Frame an Energy-Storage Dispatch Decision as a Bayesian Decision Problem
Overview
What this challenge is about.
You receive 2 years of hourly spot-price data, 2 years of wind generation data, and a manufacturer's battery degradation model. Frame dispatch as a Bayesian decision problem: model price one-hour-ahead with a posterior, action = charge/hold/discharge, reward = revenue minus degradation cost. Implement a one-step-ahead myopic Bayesian policy and (optionally) a finite-horizon look-ahead variant. Back-test on the most recent 6 months and compare to the heuristic on revenue, battery-cycle count, and worst-day PnL.
The Brief
What you'll do, and what you'll demonstrate.
Outperform the heuristic dispatch policy on revenue per cycle by framing dispatch as a Bayesian decision problem.
Earning criteria — what you'll demonstrate
- Express a real operational decision as a Bayesian decision problem
- Model utility as revenue minus degradation cost explicitly
- Build a posterior over short-horizon price and use it in decision-making
- Communicate Bayesian decision theory to a non-statistical trading team
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
Applied AI Scientist
Bayesian decision framings of real operational problems are exactly the kind of work applied AI scientists do at energy, logistics, and trading firms.
This challenge sharpens
- bayesian-decision-theory
- utility-modeling
- policy-evaluation
Data Scientist
Building a posterior on price data and using it in a back-tested policy is a high-leverage data-scientist project in energy markets.
This challenge sharpens
- price-modeling
- back-testing
- python
ML Researcher
Clean Bayesian formulations are the entry point for sequential-decision research; this challenge proves the formulation muscle.
This challenge sharpens
- bayesian-decision-theory
- utility-modeling
- policy-evaluation