vehicle-service-engine-nci 1.0.0

OOP business-rules engine for vehicle service bookings: cancellation policy, dynamic labor pricing, and bay-overbooking resolution.

vehicle-service-engine-nci

A small, dependency-free, object-oriented Python library implementing the business-rule logic for a vehicle service booking platform. Built for a cloud computing coursework project (NCI) and used inside AWS Lambda handlers, but has no AWS dependency itself, so it is fully unit-testable in isolation and reusable elsewhere.

What it does

• cancellation_policy — Strategy pattern. FlexiblePolicy, StandardPolicy, and NoShowStrictPolicy each compute a deposit refund percentage based on how many hours before the appointment a customer cancels. get_policy() is a small factory that resolves a policy name (e.g. from a database record) to an instance.
• labor_pricing — Strategy pattern. OccupancyBasedPricing raises price as service bays fill up; LastMinuteDiscountPricing discounts slots that are still empty close to the appointment date; BlendedPricing combines both. LaborPricingEngine is the façade callers use, with the strategy swappable at construction time.
• bay_allocation_resolver — Resolves overbooking: given a list of competing appointments and the number of bays actually available, decides which appointments keep their slot (highest customer tier first, then earliest booking time) and suggests compensation for the rest.

Install

pip install vehicle-service-engine-nci

Quick example

from vehicle_service_engine import LaborPricingEngine, BlendedPricing, get_policy

# Dynamic labor pricing
engine = LaborPricingEngine(BlendedPricing())
price = engine.price_for(
    base_rate=119.80,
    bay_occupancy_pct=0.8,
    days_until_appointment=1,
    is_peak_day=True,
)

# Cancellation refund calculation
policy = get_policy("STANDARD")
refund = policy.compute_refund(
    deposit_amount=50.0,
    appointment_time=some_datetime,
)

Why a separate library

Both the appointment-booking Lambda and (potentially) an admin/rescheduling tool need the same pricing and allocation rules. Keeping this logic in one importable, AWS-agnostic package means it is unit tested once, versioned independently of the Lambda deployment packages, and can be reused outside the serverless backend if needed.

License

MIT