Decision Science

MDP Sequencing & SHAP Explainability

Market entry as a Markov Decision Process, and a closed-form Shapley decomposition of the market-attractiveness score — both fully interactive.

Optimal entry sequence (MDP)

Slide the cold-start penalty and watch the optimal policy flip between US-first (value-greedy) and home-first (risk-adjusted).

Cold-start penalty0.40

1.00 = no penalty (naïve). Lower = entering US/EU/Japan cold (with zero reference logos) is much less likely to succeed.

$11.8M

Optimal-policy expected value

Optimal entry order

SEA → US → W. Europe → Japan/Korea → India

Naïve (no penalty) policy value: $16.1M, order US → W. Europe → Japan/Korea → SEA → India

Entry sequence — expected net value per step

The market score is an additive weighted model — adjust the rubric weights and watch the ranking and decomposition respond in real time.

Rubric weights

Defaults: Revenue 30% · Data-ready 20% · the rest 10–15% each.

Revenue30%

Data-ready20%

Home adv15%

Comp (inv)15%

Reg (inv)10%

Cost (inv)10%

Decision plot — what drives each market's score

Cumulative contribution of each criterion, left to right.

Ranking

Why SEA scores 4.05

Green/teal pushes the score up, red pulls it down.