Build capacity-planning model exporting head-room metrics for booking peak windows

[thlpkee20-wq]

Description

Add a capacity-planning analyzer that projects RPS, p99 latency, and resource utilization for booking-intent creation across the next N weeks based on historical traffic and known events, exporting head-room metrics.

Requirements and context

• Must be secure, tested, and documented
• Should be efficient and easy to review
• Relevant code: src/metrics/
• Output must be machine readable for autoscaler input

Suggested execution

• Fork the repo and create a branch
• git checkout -b feat/capacity-planning-model
• Implement changes
  • Add src/scripts/capacity-planner.ts
  • Pull historical metrics; project with Holt-Winters
  • Emit capacity_headroom_ratio Prometheus gauge
• Validate security and correctness assumptions

Test and commit

• Run tests
  • npm test
• Cover edge cases
  • Insufficient history must return wide confidence band, not crash
• Include test output and notes

Example commit message

feat: capacity planning head-room model

Guidelines

• Minimum 95 percent test coverage
• Clear documentation
• Timeframe: 96 hours

[Mathew2k-hash]

@Mathew2k-hash has applied to work on this issue as part of the Stellar Wave Program's 6th wave.

I’d like to work on this issue because it combines forecasting, observability, and system reliability—areas I have experience with. I have built TypeScript services that process metrics, generate operational insights, and expose machine-readable outputs for automation. My experience includes implementing statistical forecasting models, designing Prometheus metrics, handling edge cases gracefully, and delivering well-tested, documented solutions. I’m confident in building a capacity-planning analyzer that provides accurate projections, actionable headroom metrics, and reliable autoscaler inputs while maintaining high test coverage and code quality.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @Mathew2k-hash to this issue.