# Innovation Hub Youth Placement Simulator

## PURPOSE
A discrete-event simulation (SimPy + NumPy) that models how fast the Ratan Tata
Innovation Hub can move Kurnool youth from enrolment into employment given limited
trainers, seats, and employer slots. It answers capacity questions the video's
employment assurance implicitly raises: how long is the queue, where is the
bottleneck, and how many trainees actually reach placement in 90 days?

## WHY THIS MATTERS
The transcript reports a promise of "fully supported employment opportunities"
for local youth. Real programmes fail on capacity, not intent: too few training
seats or employer slots create invisible backlogs. Simulating the pipeline before
committing resources is exactly the engineering discipline that keeps a public
assurance honest.

## HOW TO BUILD
Model each stage as a SimPy resource, then run many seeded replications.

```python
import simpy

def trainee(env, name, training, placement, stats):
    with training.request() as req:
        yield req
        yield env.timeout(env.rng.uniform(10, 20))   # days of training
    with placement.request() as req:
        yield req
        wait = env.now
        yield env.timeout(env.rng.uniform(5, 15))    # days to place
    stats["placed"] += 1
```

1. Create `training = env.Resource(env, capacity=SEATS)` and a placement resource.
2. Spawn trainees as a Poisson-ish arrival process with a seeded RNG for repeatability.
3. Track waiting time per stage and total time-to-placement.
4. Sweep seats/placements across scenarios and report mean queue length.

## WHEN TO USE IN PRODUCTION
Use it to size training batches, justify budget for trainers, or plan employer
partnerships before a placement drive. Do NOT use simulation as a forecast of
individual outcomes — it models aggregate capacity. Avoid it when you already have
enough historical throughput data to fit a simple queueing formula.

## HOW IT HELPS STUDENTS
Students learn discrete-event thinking, SimPy resources and processes, seeded
reproducibility, and NumPy-driven statistical summarisation — a rare, employable
skill set that goes far beyond CRUD tutorials.

## STARTER CODE
`starter/main.py` builds the two-stage SimPy pipeline with a seeded NumPy RNG,
runs scenario sweeps, and prints mean time-to-placement and queue statistics, with
`# TODO` markers for dropout and multi-skill routing.
