Design Cab Booking (LLD)

1. Requirements

• Functional

  • Riders request trips with pickup and dropoff; system matches or assigns a driver vehicle.
  • Track trip lifecycle: requested, assigned, in progress, completed, cancelled.
  • Rider and driver both see trip state; fare estimate from distance and surge rules.

• Non-Functional

  • No double-assignment of one driver to concurrent active trips.
  • Pluggable matching (nearest, highest rating) and pricing.

• Assumptions / Out of Scope

  • Maps routing is external; distances provided by stub or API.

2. Core Entities

3. Class Diagram

4. State / Sequence Diagram (where relevant)

5. Design Patterns Applied

• Strategy — Matching and surge pricing policies. Strategy pattern.
• Observer — Trip state notifications to rider and driver apps. Observer pattern.
• State — Trip state machine. State pattern.

6. Implementation

Go

package cab

type Location struct {
    Latitude, Longitude float64
}

type Trip struct {
    Identifier string
    RiderID    string
    DriverID   string
    State      string
    Pickup     Location
    Dropoff    Location
}

type TripBookingService struct {
    Matching MatchingService
    Fare     FareEstimator
}

func (service *TripBookingService) RequestTrip(riderID string, pickup, dropoff Location) (*Trip, error) { /* ... */ }

JavaScript

class Trip {
  constructor({ identifier, riderId, pickup, dropoff }) {
    this.identifier = identifier;
    this.riderId = riderId;
    this.driverId = null;
    this.state = 'requested';
    this.pickup = pickup;
    this.dropoff = dropoff;
  }
}

class TripBookingService {
  constructor({ matchingService, fareEstimator }) {
    this.matchingService = matchingService;
    this.fareEstimator = fareEstimator;
  }
  requestTrip({ riderId, pickup, dropoff }) { /* ... */ }
}

7. Concurrency / Thread Safety

• Collisions: Two riders matched to same nearest driver; driver accepts overlapping trips.
• Granularity: Compare-and-set on driver status; transactional trip creation with row locks on driver pool.
• Go: Mutex per driver id during assignment; idempotent RequestTrip with client token.

8. Extensibility & Followups

• Shared rides with multiple waypoints as composite trip graph.
• Driver earnings ledger and weekly payout export.
• Edge cases: rider cancel after assign, no-show fees, safety escalation flags.