fleetmanager.model.qampo package

Submodules

fleetmanager.model.qampo.classes module

This file defines the different classes used in the solution.

class fleetmanager.model.qampo.classes.AlgorithmParameters(*, time_limit_in_seconds: int = None)[source]

Bases: BaseModel

time_limit_in_seconds: Optional[int]

Defines for how long, an algorithm is allowed to be run.

class fleetmanager.model.qampo.classes.AlgorithmType(value)[source]

Bases: Enum

What type of algorithm to be used.

EXACT_CP = 'exact_cp'

Exact constraint programming algorithm.

EXACT_MIP = 'exact_mip'

Exact mixed-integer programming algorithm.

GREEDY = 'greedy'

Greedy algorithm.

class fleetmanager.model.qampo.classes.Assignment(*, vehicle: BaseVehicle, route: Trips, variable_cost: float, co2_emission_in_tons: float)[source]

Bases: BaseModel

An assignment is trips assigned to a specific vehicle.

co2_emission_in_tons: float

The CO2 emission of the route measured in tons.

route: Trips

Contains a list of trips. A new trip can start immediately, after the previous trip ends, thus, ending at 08:00 means, the next step can start at 08:00.

variable_cost: float

The variable cost for the route.

vehicle: BaseVehicle

The vehicle, a route is assigned to.

class fleetmanager.model.qampo.classes.BaseVehicle(*, variable_cost_per_kilometer: float, co2_emission_gram_per_kilometer: float)[source]

Bases: BaseModel

A base vehicle consists of a variable cost of kilometers and an amount of CO2 emission per kilometer measured in grams.

class Config[source]

Bases: object

allow_mutation = False
co2_emission_gram_per_kilometer: float

The CO2 emission per kilometer measured in grams.

variable_cost_per_kilometer: float

The variable cost per kilometer.

class fleetmanager.model.qampo.classes.Fleet(*, vehicles: list[fleetmanager.model.qampo.classes.Vehicle], employee_car: BaseVehicle, emission_cost_per_ton_co2: float = None)[source]

Bases: BaseModel

A tradeoff between C02 and money for a list of vehicles.

Official documents say, somewhere between 1500 and 5000 dkk/ton is an optional list for possible parameters for scenarios, otherwise, default values are set.

emission_cost_per_ton_co2: Optional[float]

The CO2 emission cost per ton. Default value will be 1000 dkk.

employee_car: BaseVehicle

An employee car.

vehicles: list[fleetmanager.model.qampo.classes.Vehicle]

A list of vehicles.

class fleetmanager.model.qampo.classes.RoutePlan(*, assignments: list[fleetmanager.model.qampo.classes.Assignment], employee_car: Assignment, total_cost: float, total_co2_emission_in_tons: float)[source]

Bases: BaseModel

A route plan consists of a list of assignments, total cost and total CO2 emission.

assignments: list[fleetmanager.model.qampo.classes.Assignment]

A list of assignments.

employee_car: Assignment

The assignment for the employee car(s)

total_co2_emission_in_tons: float

The total amount of CO2 emission of the assignments measured in tons.

total_cost: float

The total cost of the routes.

class fleetmanager.model.qampo.classes.RoutingAlgorithm[source]

Bases: ABC

An ABC for doing a single day routing assignment.

abstract optimize_single_day(trips: list[fleetmanager.model.qampo.classes.Trip], vehicles: list[fleetmanager.model.qampo.classes.Vehicle], employee_car: BaseVehicle, emission_cost_per_ton_co2: float = 1500, time_limit_in_seconds: int = 60) → RoutePlan[source]
class fleetmanager.model.qampo.classes.Trip(*, id: int, start_time: datetime, end_time: datetime, length_in_kilometers: float)[source]

Bases: BaseModel

A trip consist of a start time, an end time and the length of the trip measured in kilometers.

The starting location and ending location are assumed to be identical for this application.

class Config[source]

Bases: object

allow_mutation = False
end_time: datetime

The time for which the trip ends.

get_trip_length_in_minutes() → int[source]

Get the trip duration in minutes (rounded up) of the trip @:return duration in number of minutes (rounded up)

id: int

A unique id for the trip.

length_in_kilometers: float

The length of the trip measured in kilometers.

start_time: datetime

The time for which the trip begins.

class fleetmanager.model.qampo.classes.Trips(*, trips: list[fleetmanager.model.qampo.classes.Trip])[source]

Bases: BaseModel

A list of trips.

trips: list[fleetmanager.model.qampo.classes.Trip]

A list of trips.

class fleetmanager.model.qampo.classes.Vehicle(*, variable_cost_per_kilometer: float, co2_emission_gram_per_kilometer: float, id: int, range_in_kilometers: float, maximum_driving_in_minutes: int, name: str = '')[source]

Bases: BaseVehicle

A vehicle consists of a range of kilometers, a variable cost of kilometers, an amount of CO2 emission per kilometer measured in grams, and a name.

class Config[source]

Bases: object

allow_mutation = False
id: int

A unique id for the vehicle.

maximum_driving_in_minutes: int

The number of minutes the vehicle can drive. Used to control e.g. electric cars recharge time in lieu of a proper battery management implementation e.g. an uptime of 16*60 minutes could model an 8 hour recharging time.

name: str

The name of the car, e.g. Toyota Yaris No. 1. Not used in the algorithm, but for easier debugging and information.

range_in_kilometers: float

The range in kilometers.

fleetmanager.model.qampo.cost_calculator module

This file defines functions for calculating the CO2 emission and costs for a route plan.

fleetmanager.model.qampo.cost_calculator.calculate_co2_emission_cost_per_kilometer_for_vehicle(vehicle: BaseVehicle, emission_cost_per_ton_co2: float) → float[source]

Calculates the CO2 emission cost per kilometer for a given vehicle.

fleetmanager.model.qampo.cost_calculator.calculate_co2_emission_gram_of_trips(vehicle: BaseVehicle, trips: list[fleetmanager.model.qampo.classes.Trip]) → float[source]

Calculates the CO2 emission (grams) of all trips combined.

fleetmanager.model.qampo.cost_calculator.calculate_co2_emission_ton_of_trips(vehicle: BaseVehicle, trips: list[fleetmanager.model.qampo.classes.Trip]) → float[source]

Calculates the CO2 emission (tons) of all trips combined.

fleetmanager.model.qampo.cost_calculator.calculate_total_length_of_trips(trips: list[fleetmanager.model.qampo.classes.Trip]) → float[source]

Calculates the length of all trips combined.

fleetmanager.model.qampo.cost_calculator.calculate_variable_cost_of_trips(vehicle: BaseVehicle, trips: list[fleetmanager.model.qampo.classes.Trip]) → float[source]

Calculates the cost of all trips combined.

fleetmanager.model.qampo.exceptions module

This file defines a number of custom exceptions.

exception fleetmanager.model.qampo.exceptions.MultipleDaysNotSupported[source]

Bases: Exception

Thrown, when multiple days are not supported (usually in the optimization).

exception fleetmanager.model.qampo.exceptions.NoSolutionFoundException[source]

Bases: Exception

Thrown, when no solution can be found.

exception fleetmanager.model.qampo.exceptions.NumberOfTripsPlannedMismatchException[source]

Bases: Exception

Thrown, when number of trips does not match number of trips assigned.

exception fleetmanager.model.qampo.exceptions.RangeExceededException[source]

Bases: Exception

Thrown, when a vehicle is assigned more kilometers than its range.

exception fleetmanager.model.qampo.exceptions.UnsupportedAlgorithmType[source]

Bases: Exception

Thrown, when the provided algorithm is not supported.

fleetmanager.model.qampo.helper_functions module

This file defines various helper functions used throughout the code.

fleetmanager.model.qampo.helper_functions.prioritize_vehicles_according_to_weighted_variable_costs(vehicles: list[fleetmanager.model.qampo.classes.BaseVehicle], emission_cost_per_ton_co2: float) → list[fleetmanager.model.qampo.classes.BaseVehicle][source]

This function weights the variable costs and CO2 emission of the vehicles and returns them in a sorted list in ascending order. :param vehicles: Vehicles to be listed. :param emission_cost_per_ton_co2: The cost of CO2 emission in dkk (or at least same currency as the variable cost on the vehicles). :return: Prioritized list of vehicles with the cheapest vehicle in position 0, second cheapest in position 1, etc.

fleetmanager.model.qampo.instance_generator module

This file defines functions for generating random test instances.

fleetmanager.model.qampo.instance_generator.generate_trips(earliest_start_hour: int, latest_start_hour: int, number_of_trips: int, minimum_length_in_kilometers: float, maximum_length_in_kilometers: float) → Trips[source]

Generates the specified number of trips with a starting time within the earliest start hour and latest start hour within the specified range in kilometers.

In general a uniform distribution is applied to the different fields.

The trips will be generated to start at minute 00/60, 15, 30 or 45.

Parameters

  • earliest_start_hour – E.g. 6 meaning, the earliest start time would be 6:00. Must be >= 0.

  • latest_start_hour – E.g. 23 meaning, the latest start time would be 23:00. Must be <= 23.

  • number_of_trips – E.g. 100 meaning, 100 trips are generated.

  • minimum_length_in_kilometers – The minimum length of a trip.

  • maximum_length_in_kilometers – The maximum length of a trip.

Returns

A class of Trips.

fleetmanager.model.qampo.main module

This file defines the main functionality of the repository - a webservice that can be called to optimize routes a day/a fleet over a couple of days. FastAPI is used for handling different HTTP requests, for instance to post an optimization.

async fleetmanager.model.qampo.main.optimize_single_day(fleet: Fleet, trips: list[fleetmanager.model.qampo.classes.Trip], algorithm_type: AlgorithmType, algorithm_parameters: Optional[AlgorithmParameters] = AlgorithmParameters(time_limit_in_seconds=60))[source]

fleetmanager.model.qampo.qampo_simulation module

fleetmanager.model.qampo.qampo_simulation.optimize_single_day(fleet: Fleet, trips: list[fleetmanager.model.qampo.classes.Trip], algorithm_type: AlgorithmType, algorithm_parameters: Optional[AlgorithmParameters] = AlgorithmParameters(time_limit_in_seconds=60))[source]

fleetmanager.model.qampo.routeplan_factory module

This file defines route plan factory to be used for classes.py.

fleetmanager.model.qampo.routeplan_factory.route_plan_from_vehicle_trip_map(vehicle_assignments: {list[fleetmanager.model.qampo.classes.Trip], <class 'fleetmanager.model.qampo.classes.Vehicle'>}) → RoutePlan[source]

Creates a route plan from a dictionary of vehicle as key and list of trips as value. :param vehicle_assignments: Assignments for vehicle. :return: Route plan.

fleetmanager.model.qampo.routeplan_tester module

This file holds basic information for testing a solution for the small trips and 3 vehicles problem.

fleetmanager.model.qampo.routeplan_tester.check_solution_for_problem(number_of_trips: int, solution: RoutePlan) → None[source]

This function performs a number of checks to find possible issues and raises exceptions if any exists. @param number_of_trips Number of trips. @param solution The solution to be checked for issues.

fleetmanager.model.qampo.routing_cp module

This file contains code, that optimizes a single day routing given a fixed fleet in an optimal manner using the Constraint programming solver from ortools.

class fleetmanager.model.qampo.routing_cp.RoutingCp[source]

Bases: RoutingAlgorithm

optimize_single_day(trips: list[fleetmanager.model.qampo.classes.Trip], vehicles: list[fleetmanager.model.qampo.classes.Vehicle], employee_car: BaseVehicle, emission_cost_per_ton_co2: float = 1500, time_limit_in_seconds: int = 60) → RoutePlan[source]

This is an exact SAT algorithm that assigns trips to vehicles based on a weight of the variable cost per kilometer and the CO2 emission. time_limit_in_seconds specifies for how long, the algorithm is allowed to be run. :param trips: List of trips in the route plan. :param vehicles: List of vehicles in the route plan. :param employee_car: Employee car a trip can be assigned to. :param emission_cost_per_ton_co2: CO2 emission cost per ton for the entire route plan. :param time_limit_in_seconds: Time limit for the running time of the algorithm. :return Routing plan created after optimization has been performed.

fleetmanager.model.qampo.routing_greedy module

This file contains code that optimizes a single day routing given a fixed fleet in a greedy manner. This is conceptually the same as the algorithm in the overall project (IFFK) that simulates car assignment previously.

class fleetmanager.model.qampo.routing_greedy.RoutingGreedy[source]

Bases: RoutingAlgorithm

optimize_single_day(trips: list[fleetmanager.model.qampo.classes.Trip], vehicles: list[fleetmanager.model.qampo.classes.Vehicle], employee_car: BaseVehicle, emission_cost_per_ton_co2: float = 1500, time_limit_in_seconds: int = 60) → RoutePlan[source]

This is a simple greedy algorithm that assigns trips to vehicles based on a weight of the variable cost per kilometer and the CO2 emission. Greedily means that earliest trips are assigned first to the cheapest vehicle. time_limit_in_seconds specifies for how long, the algorithm is allowed to be run. :param trips: List of trips in the route plan. :param vehicles: List of vehicles in the route plan. :param employee_car: Employee car a trip can be assigned to. :param emission_cost_per_ton_co2: CO2 emission cost per ton for the entire route plan. :param time_limit_in_seconds: Time limit for the running time of the algorithm. :return Routing plan created after optimization has been performed.

fleetmanager.model.qampo.routing_mip module

This file contains code that optimizes a single day routing given a fixed fleet in an optimal manner using the MIP solver from ortools (utilizing SCIP).

class fleetmanager.model.qampo.routing_mip.RoutingMip[source]

Bases: RoutingAlgorithm

optimize_single_day(trips: list[fleetmanager.model.qampo.classes.Trip], vehicles: list[fleetmanager.model.qampo.classes.Vehicle], employee_car: BaseVehicle, emission_cost_per_ton_co2: float = 1500, time_limit_in_seconds: int = 60) → RoutePlan[source]

An exact MIP algorithm that will assign trips to vehicles based on a weight of the variable cost per kilometer and the CO2 emission. time_limit_in_seconds specifies for how long, the algorithm is allowed to be run. :param trips: List of trips in the route plan. :param vehicles: List of vehicles in the route plan. :param employee_car: Employee car a trip can be assigned to. :param emission_cost_per_ton_co2: CO2 emission cost per ton for the entire route plan. :param time_limit_in_seconds: Time limit for the running time of the algorithm. :return Routing plan created after optimization has been performed.

fleetmanager.model.qampo.validation module

This file has functions for validating input data etc.

fleetmanager.model.qampo.validation.check_trips_only_has_single_date(trips: list[fleetmanager.model.qampo.classes.Trip]) → None[source]

Checks if only one single start date is present in the list, or else raises an MultipleDaysNotSupported. :param trips: Trips.

Module contents