Vehicle data model used in optimization API

agent_idstring<uuid>required

Represents and validates UUID string

latnumberrequired

Current latitude of the vehicle

lonnumberrequired

Current longtitude of the vehicle

assigned_nodes object[]

A list of nodes that have been assigned to a vehicle. If this data is included in the request, the optimization engine will assume that these nodes have already been assigned to a vehicle. This can be an outcome of previous calculations.

Array [

uidstring<uuid>required

Unique identifier of the node

booking_uidstring<uuid>

Required for pickup and dropoff nodes. Unique identifier of the booking this node belongs to (used to correlate pickup and dropoff nodes).

stop_idstring

Transit stop corresponding to this node.

latnumberrequired

Latitude of the node

lonnumberrequired

Longtitude of the node

demand objectrequired

The number of ordinary passengers to be transferred or a dictionary where the key is the type of seat (ordinary passenger is "passenger") and the value is the number of seats required.

In Logistics this could be how many tons of goods the vehicle need to pickup/dropoff from this node

property name*integer

open_time_tsstring<date-time>required

Earliest date and time when the node processing can be started.

close_time_tsstring<date-time>required

Latest date and time when the node processing can be started.

service_timenumberrequired

Time required for a vehicle to be serviced (loaded or unloaded) at this node

node_typeNode types enum (string)required

Sets type of the node

Possible values: [dropoff, point, pickup, vehicle_position, depot]

close_time_ts_dynamicstring<date-time>

Latest date and time when the node processing can be started. Usually equals close_time_ts and is applied to the dynamic algorithm type.

location_namestring

Name of the location.

location_codestring

Code of the location if applicable.

scheduled_tsstring<date-time>

Date and time when this node is supposed to be visited. NOTE: This should be omitted unless required explicitly by the use case.

max_trip_durationnumber

Dropoff node's close_time_ts_dynamic minus pickup node's open_time_ts_dynamicif algorithm is dynamic Dropoff node's close_time_ts minus pickup node's open_time_ts otherwise

started_service_at_tsstring,null<date-time>nullable

Date and time when node service started. Is required for vehicle.completed_nodes, otherwise may be omitted.

lifo_order_checkboolean

Flag that enables LIFO order check for nodes of this booking. Used only if use_lifo_order_check is set for model. Enables or disables the LIFO order logic for this node, LIFO logic is a soft constraint forcing the pickups and drop offs to happen in LIFO sequence related to the vehicles (last picked up order is prioritized for the drop off)

Default value: false

lifo_order_penaltynumber | nullnullable

Penalty value that defines that LIFO constraint for this node is soft.

groupsstring[]

A list of group names as strings for this node. Items in this list are used for processing the mutually-exclusive groups constraint.

max_slacknumber

maximum slack limitation for this node.

weightnumber | nullnullable

Transit stop weight. Weight of the node that is added to cost function if the stop is active.

allow_jumpboolean

Whether to allow the node to be re-assigned to another vehicle.

penaltynumber

Rejection penalty for the node (a large value added to the overall cost function) for not performing this booking. If the value is null, the penalty value is taken from model_parameters.booking_penalty settings. This value should be large enough to justify the cost of traveling to the node or the cost of violating soft constraints, if any, to have the node assigned to a vehicle. Values can vary from hundreds to millions depending on the use case.

matrix_timestampstring,null<date-time>nullable

Timestamp for which transits from this node should be calculated. Default value is node.open_time_ts

geofence_idstring | nullnullable

Geofence id for this node.

geofence_idsstring[]

List of geofence ids for this node (if node is close to geofences border).

trip_costnumber<float>

Price of trip time for this booking. The overall trip cost for a booking is calculated as real travel time * trip_cost. The default value is defined in the scheduler settings.

finalization_typestring

Defines options to get the final scheduled time of this node. If set to 'max', the algorithm tries to "shift" a solution to schedule the node visit as late as possible within a given solution (thus minimizing slack); when set to 'min', it tries to schedule the node visit as early as possible. This is a post-processing step that does not impact the solution but tries to adjust the result.

Possible values: [min, max]

vehicle_characteristics object

Required vehicle characteristics to perform this node. Required vehicle characteristics for booking must be a dictionary mapping the required characteristic name to characteristic minimum value (as value convertible to integer) or value range (as dict, containing 'min' and 'max' keys).

Important note: all characteristics must be present in the list of characteristics for all vehicles.

property name* object

anyOf

number

number

Value of the characteristic, for example:

{
    "manpower": 3
}

Ranged characteristic

minnumber

Minimum value of the characteristic

maxnumber

Maximum value of the characteristic

boolean

boolean

vehicle_labelsobject | nullnullable

Vehicle labels contain constraints, represented as dictionaries, that may contain string labels or sub-constraints. A constraint is a dictionary containing only one entry with keys: "and", "or", or "not". The keys "and" and "or" define the AND or OR logic functions, and their values should be a list of either string labels or dictionaries containing sub-constraints. The key "not" defines the NOT logic function, and its value can be a single string or a single sub-constraint.

Example of vehicle labels constraints: '1' AND ('2' OR '3') AND NOT ('4' AND '5') as:

{"and": [{"or": ["2", "3"]}, {"not": {"and": ["4", "5"]}}]}

end_of_tripboolean

The end-of-vehicle-trip marker must be set for all nodes located at the depot(s). This allows the scheduler to calculate the number of vehicle trips and obey the number_of_trips constraint.

time_windowsstring[]

List of additional time windows.

]

completed_nodes object[]

A list of the nodes that have been completed by a vehicle. In case of having this data in the request, the optimization engine will assume that these nodes have been already completed by a vehicle. THis is required to respect cumulative constraints such as maximum travelled distance. This can be a result of previous calculations.

Array [

uidstring<uuid>required

Unique identifier of the node

booking_uidstring<uuid>

Required for pickup and dropoff nodes. Unique identifier of the booking this node belongs to (used to correlate pickup and dropoff nodes).

stop_idstring

Transit stop corresponding to this node.

latnumberrequired

Latitude of the node

lonnumberrequired

Longtitude of the node

demand objectrequired

The number of ordinary passengers to be transferred or a dictionary where the key is the type of seat (ordinary passenger is "passenger") and the value is the number of seats required.

In Logistics this could be how many tons of goods the vehicle need to pickup/dropoff from this node

property name*integer

open_time_tsstring<date-time>required

Earliest date and time when the node processing can be started.

close_time_tsstring<date-time>required

Latest date and time when the node processing can be started.

service_timenumberrequired

Time required for a vehicle to be serviced (loaded or unloaded) at this node

node_typeNode types enum (string)required

Sets type of the node

Possible values: [dropoff, point, pickup, vehicle_position, depot]

close_time_ts_dynamicstring<date-time>

Latest date and time when the node processing can be started. Usually equals close_time_ts and is applied to the dynamic algorithm type.

location_namestring

Name of the location.

location_codestring

Code of the location if applicable.

scheduled_tsstring<date-time>

Date and time when this node is supposed to be visited. NOTE: This should be omitted unless required explicitly by the use case.

max_trip_durationnumber

Dropoff node's close_time_ts_dynamic minus pickup node's open_time_ts_dynamicif algorithm is dynamic Dropoff node's close_time_ts minus pickup node's open_time_ts otherwise

started_service_at_tsstring,null<date-time>nullable

Date and time when node service started. Is required for vehicle.completed_nodes, otherwise may be omitted.

lifo_order_checkboolean

Flag that enables LIFO order check for nodes of this booking. Used only if use_lifo_order_check is set for model. Enables or disables the LIFO order logic for this node, LIFO logic is a soft constraint forcing the pickups and drop offs to happen in LIFO sequence related to the vehicles (last picked up order is prioritized for the drop off)

Default value: false

lifo_order_penaltynumber | nullnullable

Penalty value that defines that LIFO constraint for this node is soft.

groupsstring[]

A list of group names as strings for this node. Items in this list are used for processing the mutually-exclusive groups constraint.

max_slacknumber

maximum slack limitation for this node.

weightnumber | nullnullable

Transit stop weight. Weight of the node that is added to cost function if the stop is active.

allow_jumpboolean

Whether to allow the node to be re-assigned to another vehicle.

penaltynumber

Rejection penalty for the node (a large value added to the overall cost function) for not performing this booking. If the value is null, the penalty value is taken from model_parameters.booking_penalty settings. This value should be large enough to justify the cost of traveling to the node or the cost of violating soft constraints, if any, to have the node assigned to a vehicle. Values can vary from hundreds to millions depending on the use case.

matrix_timestampstring,null<date-time>nullable

Timestamp for which transits from this node should be calculated. Default value is node.open_time_ts

geofence_idstring | nullnullable

Geofence id for this node.

geofence_idsstring[]

List of geofence ids for this node (if node is close to geofences border).

trip_costnumber<float>

Price of trip time for this booking. The overall trip cost for a booking is calculated as real travel time * trip_cost. The default value is defined in the scheduler settings.

finalization_typestring

Defines options to get the final scheduled time of this node. If set to 'max', the algorithm tries to "shift" a solution to schedule the node visit as late as possible within a given solution (thus minimizing slack); when set to 'min', it tries to schedule the node visit as early as possible. This is a post-processing step that does not impact the solution but tries to adjust the result.

Possible values: [min, max]

vehicle_characteristics object

Required vehicle characteristics to perform this node. Required vehicle characteristics for booking must be a dictionary mapping the required characteristic name to characteristic minimum value (as value convertible to integer) or value range (as dict, containing 'min' and 'max' keys).

Important note: all characteristics must be present in the list of characteristics for all vehicles.

property name* object

anyOf

number

number

Value of the characteristic, for example:

{
    "manpower": 3
}

Ranged characteristic

minnumber

Minimum value of the characteristic

maxnumber

Maximum value of the characteristic

boolean

boolean

vehicle_labelsobject | nullnullable

Vehicle labels contain constraints, represented as dictionaries, that may contain string labels or sub-constraints. A constraint is a dictionary containing only one entry with keys: "and", "or", or "not". The keys "and" and "or" define the AND or OR logic functions, and their values should be a list of either string labels or dictionaries containing sub-constraints. The key "not" defines the NOT logic function, and its value can be a single string or a single sub-constraint.

Example of vehicle labels constraints: '1' AND ('2' OR '3') AND NOT ('4' AND '5') as:

{"and": [{"or": ["2", "3"]}, {"not": {"and": ["4", "5"]}}]}

end_of_tripboolean

The end-of-vehicle-trip marker must be set for all nodes located at the depot(s). This allows the scheduler to calculate the number of vehicle trips and obey the number_of_trips constraint.

time_windowsstring[]

List of additional time windows.

]

partial_routestring<uuid>[]

The list containing UIDs of nodes which represents the vehicle's partial route. The order in the list must represent the actual sequence of assigned nodes to the vehicle.

capacity objectrequired

Represents capacity constrains of the vehicle. This is a free form dictinoary with keys that must match with relevant demand values for nodes, otherwise the optimization will not be able to solve the VRP problem.

property name*integer

start_timestring | nullnullable

Timestamp of the vehicle operations start time.

end_timestring | nullnullable

Timestamp of the vehicle operations end time.

vehicle_costinteger

Vehicle cost factor.

dynamic_break_durationinteger

Allowed duration of the vehicle break (i.e., driver break) in seconds. The break node is selected from one of the nodes in the path with the greatest slack. The break node cannot be the first or the last in the path. If slacks are identical, the break node is selected to be as close as possible to the middle of the path in prebook_cvrptw mode or to have minimal possible demand in prebook mode.

dynamic_break_min_path_durationinteger

Minimum path duration when dynamic breaks are required. If real path duration is less than this value, then no dynamic break will be added. "Break" is added only if the vehicle's path is longer than dynamic_break_min_path_duration. Setting dynamic_break_min_path_duration makes breaks always added. Break is not added if there is only one node in the path.

max_dropoff_locationsnumber

Maximum number of consecutive, non-identical drop-off locations after the last pickup location.

labelsstring[]nullable

List of vehicle lables. Use to select approriate bookings for assignments based on rules defined in a node.vehicle_labels field.

zero_cost_if_only_partial_routesboolean

If set to true, vehicle cost for this vehicle will not be added to objective value if the route for this vehicle contains only partial_route + partial_route_end

number_of_tripsinteger

Maximum number of times a vehicle can return to the depot(s). Nodes located at a depot must be marked with end_of_trip=true. In CVRPTW optimization mode (with all orders picked up at a single location/warehouse), this parameter is ignored since only one trip per vehicle is supported.

max_trip_durationinteger

Max trip duration in seconds. New bookings are assigned to the vehicle as long as the total trip does not exceed this value. If the property max_trip_duration is not set explicitly, it is calculated as:

• for dynamic algorythm Dropoff node's close_time_ts_dynamic minus pickup node's open_time_ts_dynamic
• in other cases Dropoff node's close_time_ts minus pickup node's open_time_ts

efficiency vehicle_efficiency (object,null)

constraints object[]

This feature enables efficiency constraints for vehicles, managing how much capacity they utilize. It works by defining a list of constraints for various demand types. These constraints focus on two optional checks during a vehicle's trip:

Minimum Load: Ensures the vehicle is loaded with at least a specified minimum capacity throughout the trip. This promotes efficient vehicle usage. Maximum Remaining Load: Ensures the vehicle's remaining load at the end of a trip doesn't exceed a specified maximum. This discourages returns to the depot with significant unused capacity. Each check can be either strict or weak:

Strict: These constraints must be met. Failing a strict constraint is unacceptable, even if it means no solution is found. Weak: These constraints are preferred but not mandatory. If a weak constraint is violated, a penalty is added to the overall objective value. The penalty for violating a weak constraint is calculated as the sum of two components:

Fixed Penalty: A pre-defined penalty value. Linear Penalty: Calculated by multiplying the cost of over/under capacity by the amount of over/under capacity.

Array [

demand_namestring

Demand name to apply the constraint. This demand type must be present in at least one full booking. If the demand set here is not present in any of the booking as a set demand type, the optimizer will fail the validation.

loaded_min_capacitynumber | nullnullable

Minimum capacity after a series of consecutive pickups before dropoffs. If check is not strict then penalty is added to cost if capacity is less than minimum. If check is strict then route with too small capacity is invalid.

loaded_min_capacity_percentagenumber | nullnullable

The same as loaded_min_capacity but represented as percentage.

loaded_undercapacity_penaltyinteger

Fixed penalty of undercapacity.

Default value: 0

loaded_undercapacity_costinteger

Cost of undercapacity. The additional penalty is calculated as the product of undercapacity_cost and the undercapacity value.

Default value: 0

loaded_undercapacity_check_strictboolean

Flag indicates that check of undercapacity must be strict.

Default value: false

unloaded_max_capacitynumber | nullnullable

Maximum capacity after a series of consecutive dropoffs before pickups. If check is not strict then penalty is added to cost if capacity is greater than maximum. If check is strict then route with too large capacity is invalid.

unloaded_max_capacity_percentagenumber | nullnullable

Same as unloaded_max_capacity but represented as percentage.

unloaded_overcapacity_penaltyinteger

Fixed penalty of overcapacity.

Default value: 0

unloaded_overcapacity_costinteger

Cost of overcapacity. The additional penalty is calculated as the product of overcapacity_cost and the overcapacity value.

Default value: 0

unloaded_overcapacity_check_strictboolean

Flag, indicates, that check for overcapacity must be strict.

Default value: false

]

partial_route_endstring

The list containing uids of nodes which represents vehicle's partial route end.

pathstring

Current vehicle's path as a sequence of (lat, lon) in encoded polyline format. Can be omitted unless required explicitly.

should_wait_untilstring<date-time>

If defined, vehicle should stay at the same location until this time is reached.

current_timestring

Current date and time (vehicle's properties are valid this time). Can be normally omitted unless required explicitly for some retrospective or in the future calculations.

routing_engine object

Used only if mixed_fleet is set to true

routing_engine_namestringrequired

Name of the engine to be used.

Possible values: [euclidian, euclidian_geo, asteria, graphhopper, mapbox, valhalla, osrm, osrme, google, here, spheroid, spheriodtd, tomtom]

Default value: asteria

road_networkstringrequired

Road network to be used by the routing engine (routing profile)

Default value: osrme

keystringrequired

API key for the mapping service

Default value: null

urlstringrequired

URL of the routing engine to request the data from

Default value: http://mapbox-osrm-proxy

curbboolean

Sets approacch values in the resuts to unrestricted with false, and curb with true value. The approaches are set only for the first and the last stop in the trip

Default value: null

intermediate_curbboolean

Sets approacch values in the resuts to unrestricted with false, and curb with true value. The approaches are set for all waypoints in a trip except for the first and the last stop

Default value: null

continue_straightboolean

Applies continue_straight flag to the routing request. WHen set to null default map setting is applied.

Default value: null

speednumber | nullnullable

Speed value used by the routing engine if fiexed speed is enabled with use_speed_in_routing flga

Default value: 0

time_factornumber

Can be used to manually adjust the time needed to move between waypoints compared to what the routing engine provides based on the vehicle profile.

Default value: 1

make_depot_zeroboolean

Default value: true

batch_matrix_sizeinteger

Defines size of the matix to request from the routing engine with a single request for optimization purposes

Default value: 250

osrme_timestamp_modestring

Defines default timestamp type to be used for requesting the route

Default value: start_time

use_speed_in_routingboolean

If enabled, forces the routing engine to use fixed speed and disregard speeds from the map

Default value: false

lifo_order_checkboolean

Flag that enables LIFO order check for this vehicle. Used only if use_lifo_order_check is set for the model.

Default value: false

max_physical_stopsinteger | nullnullable

The number of physical stops (excluding depot, points, and vehicle_positions).

geofence_idsinteger[]

Ids of geofences, that are served by this vehicle or empty list means no restrictions.

amortized_linear_cost_factorinteger

Amortized vehicle cost (linear part). The amortized cost for a vehicle during optimization is calculated as L - QN^2, where L is the linear part of the amortized cost factor, Q is the quadratic part, and N is the number of nodes in the vehicle's path.

amortized_quadratic_cost_factorinteger

Amortized vehicle cost (quadratic part)

matrix_idstring

Identifier of the matrix. If omitted or null, the routing engine will be used to obtain the matrix.

max_trip_lengthnumber<float>

Maximum trip length for this vehicle. Units are the same as the routing engine uses for distance calculation.

characteristics object

Vehicle characteristics. It must be a dictionary mapping a characteristic name to a characteristic value convertible to an integer. For every vehicle, all required characteristics (present in bookings or nodes) must be present.

property name* object

anyOf

number

number

Value of the characteristic, for example:

{
    "manpower": 3
}

Ranged characteristic

minnumber

Minimum value of the characteristic

maxnumber

Maximum value of the characteristic

boolean

boolean

dynamic_break_start_timestring<date-time>

Minimum time when a dynamic break can start. The time window for a break is set using the dynamic_break_start_time and dynamic_break_end_time fields. These fields define the time interval when a break can occur. Setting only one of these parameters (a half-window) is also a valid configuration. In this case, only the minimum or maximum time will be checked.

dynamic_break_end_timestring<date-time>

Maximum time when dynamic break should end.

dynamic_break_avg_time_between_breaksinteger

Average time between breaks in seconds. When dynamic_break_avg_time_between_breaks is not set, single break mode is used. In single break mode, there is only zero or one break per path. Dynamic breaks are modeled as some amount of time after servicing one of the nodes in the path. For every vehicle, there can be only zero or one dynamic break. Break duration sets the amount of "free" time on the break node. When dynamic_break_duration is set to zero, breaks will not be added to any nodes for this vehicle. If dynamic_break_avg_time_between_breaks is set to a non-zero value, then multiple breaks mode is used. In multiple breaks mode, there might be zero or more breaks per path. The dynamic_break_avg_time_between_breaks parameter sets the amount of time between the end of the previous break and the start of the next break. The dynamic_break_max_latency parameter sets the maximum amount of time the break can be late. There might be only one dynamic break per transit (between two nodes).

dynamic_break_max_latencyinteger

This value specifies the maximum allowable delay (lateness) for the break.

max_pickup_locationsinteger

Maximum number of consecutive, non-identical pickup locations after the last drop-off location.

first_solutionstring[]

Each string represents a node.uid. When engine_settings.solver_parameters.first_solution_strategy is set to EVALUATOR_STRATEGY, these lists assign the first solution route for the vehicles.

Vehicle data model used in optimization API

{
  "agent_id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "lat": 0,
  "lon": 0,
  "assigned_nodes": [
    {
      "uid": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
      "booking_uid": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
      "stop_id": "string",
      "lat": 0,
      "lon": 0,
      "demand": {},
      "open_time_ts": "2024-07-29T15:51:28.071Z",
      "close_time_ts": "2024-07-29T15:51:28.071Z",
      "service_time": 0,
      "node_type": "dropoff",
      "close_time_ts_dynamic": "2024-07-29T15:51:28.071Z",
      "location_name": "string",
      "location_code": "string",
      "scheduled_ts": "2024-07-29T15:51:28.071Z",
      "max_trip_duration": 0,
      "started_service_at_ts": "2024-07-29T15:51:28.071Z",
      "lifo_order_check": false,
      "lifo_order_penalty": 0,
      "groups": [
        "string"
      ],
      "max_slack": 0,
      "weight": 0,
      "allow_jump": true,
      "penalty": 0,
      "matrix_timestamp": "2024-07-29T15:51:28.071Z",
      "geofence_id": "string",
      "geofence_ids": [
        "string"
      ],
      "trip_cost": 0,
      "finalization_type": "min",
      "vehicle_characteristics": {},
      "vehicle_labels": "Unknown Type: object,null",
      "end_of_trip": true,
      "time_windows": [
        "string"
      ]
    }
  ],
  "completed_nodes": [
    {
      "uid": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
      "booking_uid": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
      "stop_id": "string",
      "lat": 0,
      "lon": 0,
      "demand": {},
      "open_time_ts": "2024-07-29T15:51:28.071Z",
      "close_time_ts": "2024-07-29T15:51:28.071Z",
      "service_time": 0,
      "node_type": "dropoff",
      "close_time_ts_dynamic": "2024-07-29T15:51:28.071Z",
      "location_name": "string",
      "location_code": "string",
      "scheduled_ts": "2024-07-29T15:51:28.071Z",
      "max_trip_duration": 0,
      "started_service_at_ts": "2024-07-29T15:51:28.071Z",
      "lifo_order_check": false,
      "lifo_order_penalty": 0,
      "groups": [
        "string"
      ],
      "max_slack": 0,
      "weight": 0,
      "allow_jump": true,
      "penalty": 0,
      "matrix_timestamp": "2024-07-29T15:51:28.071Z",
      "geofence_id": "string",
      "geofence_ids": [
        "string"
      ],
      "trip_cost": 0,
      "finalization_type": "min",
      "vehicle_characteristics": {},
      "vehicle_labels": "Unknown Type: object,null",
      "end_of_trip": true,
      "time_windows": [
        "string"
      ]
    }
  ],
  "partial_route": [
    "3fa85f64-5717-4562-b3fc-2c963f66afa6"
  ],
  "capacity": {},
  "start_time": "string",
  "end_time": "string",
  "vehicle_cost": 0,
  "dynamic_break_duration": 0,
  "dynamic_break_min_path_duration": 0,
  "max_dropoff_locations": 0,
  "labels": "Unknown Type: array,null",
  "zero_cost_if_only_partial_routes": true,
  "number_of_trips": 0,
  "max_trip_duration": 0,
  "efficiency": "Unknown Type: object,null",
  "partial_route_end": "string",
  "path": "string",
  "should_wait_until": "2024-07-29T15:51:28.071Z",
  "current_time": "string",
  "routing_engine": {
    "routing_engine_name": "asteria",
    "road_network": "osrme",
    "key": null,
    "url": "http://mapbox-osrm-proxy",
    "curb": null,
    "intermediate_curb": null,
    "continue_straight": null,
    "speed": 0,
    "time_factor": 1,
    "make_depot_zero": true,
    "batch_matrix_size": 250,
    "osrme_timestamp_mode": "start_time",
    "use_speed_in_routing": false
  },
  "lifo_order_check": false,
  "max_physical_stops": 0,
  "geofence_ids": [
    0
  ],
  "amortized_linear_cost_factor": 0,
  "amortized_quadratic_cost_factor": 0,
  "matrix_id": "string",
  "max_trip_length": 0,
  "characteristics": {},
  "dynamic_break_start_time": "2024-07-29T15:51:28.071Z",
  "dynamic_break_end_time": "2024-07-29T15:51:28.071Z",
  "dynamic_break_avg_time_between_breaks": 0,
  "dynamic_break_max_latency": 0,
  "max_pickup_locations": 0,
  "first_solution": [
    "string"
  ]
}