Works with the AI Agents you already use
demand and capacity forecasting analyst
Historical data often hides the staffing gaps that lead to service failure.
$2.99
demand and capacity forecasting analyst
Example session with this skill installed
Forecast demand for the Support-E1 dataset (approved by Ops Lead) using Erlang C for a 30s/80% target. History is in history.csv, capacity in fte.csv. Current AHT is 420s.
- Read your context and instructions
- Compiled the demand and capacity
I have processed the Support-E1 dataset. The analysis used history from 2023-01-01 to 2024-03-01 to forecast the next 8 weeks.
Demand and Capacity Forecast — Support Operations, 2024-03-15
1. Scope, sources and assumptions
- Sources: Dataset E1 (Support-E1), approved by Ops Lead.
- Grid: Weekly, 8-week horizon.
- Service Target: Erlang C (80% in 30s).
- Assumptions: AHT 420s, Shrinkage 20%, 40 paid hours per FTE period.
2. Data used and exclusions
- Periods: 60 weeks of history used.
Exclusions: 2 records excluded due to missing demand values. Person-level identifiers in source files were ignored.
3. Forecast and uncertainty
| Period | Point | Low | High | Calendar |
|---|---|---|---|---|
| 2024-W12 | 1,250 | 1,180 | 1,320 | 1.0 |
| 2024-W13 | 1,280 | 1,210 | 1,350 | 1.0 |
| 2024-W14 | 1,450 | 1,370 | 1,530 | 1.15 |
Analytical estimate — validation required
Selected Model: Triple Exponential Smoothing (Holt-Winters) due to lowest WAPE (4.2%) and Bias (0.8%). Seasonal spikes were captured within the empirical interval.
4. Capacity gaps and service risks
| Period | Demand | Required FTE | Available FTE | Gap | Utilisation | Service Level | Risk |
|---|---|---|---|---|---|---|---|
| 2024-W12 | 1,250 | 22.4 | 24.0 | 1.6 | 78% | 84% | Low |
| 2024-W13 | 1,280 | 22.9 | 22.0 | -0.9 | 87% | 72% | High |
| 2024-W14 | 1,450 | 26.1 | 22.0 | -4.1 | 94% | 45% | Critical |
Worst gap identified in 2024-W14 with a shortfall of 4.1 FTE.
5. Staffing scenarios
- Scenario A: +2 FTE in W13-W14. Result: Service level improves to 79% in W13.
- Scenario B: AHT reduction to 380s. Result: Capacity gap closed for W13.
6. Draft actions requiring approval
- Draft: Increase seasonal headcount for W14 by 4 FTE. Management decision required.
- Draft: Review AHT assumptions for W14 spike. Management decision required.
7. Limitations and items for human review
- Information gap — organizational validation required: Shrinkage is a static 20% assumption.
- The model describes patterns only and makes no causal claims regarding the W14 spike.
- This analysis supports, and does not replace, professional judgment and management's decision.
Next steps
- Validate the demand multiplier for the W14 holiday spike.
- Review the 420s AHT assumption against recent performance trends.
- Obtain approval for the proposed seasonal headcount increase.
Connects securely to your tools. The creator never sees your data.
About this skill
Historical data often hides the staffing gaps that lead to service failure. This skill quantifies future operational shortfalls by running a deterministic backtest and forecast against your actual capacity schedules.
What it does
- Forecasts demand from approved aggregate history using a rolling-origin backtest to handle seasonal spikes.
- Calculates staffing requirements by converting demand into FTE needs via utilisation rules or Erlang C queue approximations.
- Identifies capacity gaps by comparing required staff against your existing capacity calendar to find shortfalls.
- Generates staffing scenarios to test "what-if" changes in FTE counts, shrinkage, or handling times.
- Reports uncertainty through empirical intervals and service-level risk assessments for every period.
How it works
- Data Registration: You provide aggregate demand history and capacity schedules, which the skill registers as traceable, approved datasets.
- Configuration: The skill builds a
request.jsondefining handling times, service targets, and model parameters like horizon and season length. - Execution: A deterministic Python script processes the request to select the best-performing model based on historical error metrics.
- Validation: You review the forecast envelope, including model bias, WAPE, and seasonal spike performance.
Frameworks & tools
This skill uses a self-contained Python execution environment (scripts/run.py) to perform all time-series modeling and queueing theory calculations.
Why this beats prompting it yourself
Standard prompts often hallucinate future numbers or fail to apply rigorous queueing math like Erlang C. This skill enforces a strict separation between data and calculation, ensuring every forecast is backed by a validated backtest rather than a linguistic guess.
Use cases
- Staffing shortfall analysis to identify exactly which weeks will miss service level targets.
- Budget justification by showing the impact of increased demand on required FTE.
- Seasonal peak planning to verify how previous spikes were handled and project future risks.
- Operational what-if testing to see the effect of changing shrinkage or target handling times.
Known limitations
Does not support person-level rostering, cost optimization, or multi-skill routing. Limited to daily or weekly periods; intraday interval profiles are not implemented.
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 data. No intraday or person-level scheduling. No cost optimization. Supports only daily or weekly grids. Forecasts are analytical estimates requiring human validation.
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