Works with the AI Agents you already use
demand forecasting for operations
Operations planning fails when demand spikes are treated as surprises rather than cycles.
$2.99
demand forecasting for operations
Example session with this skill installed
Forecast demand for the next 6 months using the attached history. Capacity is 200 units per month. Season length is 12. History: 2023-01: 150, 2023-02: 165... [24 months of data].
- Read your context and instructions
- Compiled the demand forecasting
The forecast has been generated using multiplicative seasonal decomposition and an OLS linear trend.
Demand Forecasting for Operations — Logistics Dept, 2024-05-20
Status: completed.
Model and assumptions
Model: Multiplicative Seasonal Decomposition with OLS Trend.
Formula: Demand = (Level + Trend * t) * Seasonal_Index + Error.
Assumptions: Seasonal patterns are stable; residuals are normally distributed.
Data sufficiency
Points: 24. Gaps: 0. Sufficiency: Pass (Supports seasonal estimation).
Seasonal indices
| Month | Index | Observations |
|---|---|---|
| Jan | 0.92 | 2 |
| Feb | 1.05 | 2 |
| Mar | 1.15 | 2 |
| ... | ... | ... |
Forecast
| Period | Point | 80% Band | 90% Band | Utilisation | Status |
|---|---|---|---|---|---|
| 2024-06 | 185 | [170, 200] | [165, 205] | 92.5% | within |
| 2024-07 | 210 | [195, 225] | [190, 230] | 105.0% | exceeds |
| 2024-08 | 195 | [180, 210] | [175, 215] | 97.5% | within |
Accuracy and uncertainty
In-sample MAPE: 4.2%. Holdout MAPE: 5.8%. Bias: -1.2%. Bands assume residual dispersion persists.
Capacity implications
Periods assessed: 6. Shortfall count: 1 (2024-07). Peak utilisation: 105.0%.
Proposed actions
- Draft: Increase processing capacity by 5% for July cycle (requires_human_approval: true).
- Draft: Shift 10 units of non-perishable backlog to August (requires_human_approval: true).
Limitations and review items
Analytical estimate — validation required. This is a heuristic baseline.
Next steps
- Review the capacity shortfall flagged for 2024-07.
- Approve or modify the proposed buffer actions.
- Verify the seasonal index for March against known promotion events.
Connects securely to your tools. The creator never sees your data.
About this skill
Operations planning fails when demand spikes are treated as surprises rather than cycles. This skill provides a traceable, seasonal demand baseline to justify capacity and stock decisions.
What it does
- Seasonal decomposition calculates centered-moving-average indices from monthly or cyclical history.
- Trend analysis applies OLS linear regression to deseasonalized data for long-term growth or decay.
- Uncertainty quantification generates 80% and 90% confidence bands based on residual dispersion.
- Accuracy reporting provides holdout MAPE and bias metrics to prevent overconfidence in the model.
- Capacity checking flags specific periods where forecasted demand exceeds your processing limits.
How it works
- Data ingestion reads time-indexed history, per-period capacity, and seasonal cycle length.
- Execution runs the
scripts/run.pyengine to perform the mathematical decomposition and OLS fit. - Validation checks data sufficiency for seasonal estimation and reports warnings for gaps.
- Drafting produces proposed operational actions that remain in a draft state for human approval.
Frameworks & tools
This skill uses a Python-based execution environment to perform statistical calculations. It relies on standard OLS and multiplicative seasonal decomposition methods for deterministic, reproducible results.
Why this beats prompting it yourself
Generic LLM prompts often hallucinate seasonal indices or fail to separate trend from cycle correctly. This skill uses a dedicated execution script to ensure every calculation is traceable, mathematically consistent, and includes rigorous holdout testing.
Use cases
- Forecast monthly returns or order volumes for the next 12 to 24 months.
- Identify upcoming months where peak seasonal demand will exceed depot capacity.
- Build a baseline seasonal model for operations reviews with a full calculation trace.
- Evaluate the reliability of current forecasts using MAPE and bias metrics.
Known limitations
Requires at least two full cycles of aggregate history for seasonal estimation. It is a heuristic baseline tool, not a model selection or auto-ML engine. Individual respondent data is not supported.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Requires aggregate history (2+ cycles). Not for workforce scheduling or model selection. Outputs are heuristic baselines, not proven optimal. Only supports even-cycle seasonal lengths (2-24).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 days ago