Management and optimization of freight delivery vehicle fleets
A method for optimizing operation of a fleet of vehicles includes providing an optimization network including a plurality of nodes and a plurality of connections therebetween. The nodes define all potential destinations in the network for freight delivery or pickup. The connections define all potential sequences of nodes traversable over the network. The method determines a pre-optimization simulation matrix including a plurality of sets of energy expenditures for traversing respective ones of each of the connections. Each of the plurality of sets of energy expenditures accounts for permutations of a plurality of different vehicle powertrains present in a fleet of vehicles. The method performs an optimization using the pre-optimization simulation matrix, fleet information indicating locations and energy states of a plurality of vehicles in the vehicle fleet, mission information indicating a pickup or delivery destination for a plurality of items of freight, and a fleet optimization objective.
1 . A method for optimizing operation of a fleet of vehicles, the method comprising:
providing a computer system-based optimization network including a plurality of nodes and a plurality of connections between respective pairs of the plurality of nodes, the plurality of nodes defining all potential destinations in the optimization network for freight delivery or pickup, the plurality of connections defining all potential sequences of nodes traversable over the optimization network;
determining with the computer system-based optimization network a pre-optimization simulation matrix including a plurality of sets of energy expenditures for traversing respective ones of each of the connections, each of the plurality of sets of energy expenditures accounting for permutations of a plurality of different vehicle powertrains present in a fleet of vehicles,
performing with the computer system-based optimization network an optimization using the pre-optimization simulation matrix, fleet information indicating current locations and current energy states of a plurality of vehicles in the vehicle fleet, mission information indicating a pickup destination and a delivery destination for a plurality of items of freight, and a fleet optimization objective, wherein the optimization determines a plurality of optimized dispatch assignments each defining at least a particular vehicle of the fleet, a selection of items from the plurality of items of freight, and a sequence of nodes for delivery or pickup the selection of items, a computational burden of the optimization performed with the computer system-based optimization network being reduced by use of the pre-optimization simulation matrix, and
dispatching in response to output of the computer system-based optimization network a plurality of vehicles of the fleet of vehicles according to the plurality of optimized dispatch assignments and controlling operation of the plurality of vehicles according to the dispatching.
2 . The method of claim 1 , wherein the plurality of different vehicle powertrains comprise two or more of a combustion engine powertrain, a combustion-electric powertrain, a purely battery-electric powertrain, and a fuel-cell electric powertrain.
3 . The method of claim 1 , wherein each set of energy expenditures further accounts for one or more of:
a first plurality of points over a range of potential initial vehicle energy states;
a second plurality of points over a range of potential initial vehicle freight load states; and
a third plurality of points over a range of potential environmental conditions.
4 . The method of claim 3 , wherein the optimization performs an interpolation between two points of at least one of the first plurality of points, the second plurality of points, and the third plurality of points, the interpolation being in response to an input indicating a value intermediate said two points.
5 . The method of claim 1 , wherein each set of energy expenditures further accounts for utilization of one or more automation features in controlling one or more vehicles of the fleet.
6 . The method of claim 1 , wherein the optimization accounts for availability of one or more energy resources in determining the plurality of optimized dispatch assignments.
7 . The method of claim 6 , wherein the one or more energy resources include one or more of a diesel fueling, natural gas fueling, battery charging, battery exchange, and hydrogen filling.
8 . The method of claim 1 , wherein the fleet optimization objective comprises a minimization of net energy expenditure of the fleet of vehicles.
9 . The method of claim 1 , comprising:
operating a machine learning component to update a model component utilized in the determining the pre-optimization simulation matrix in response to fleet operation feedback provided by a telematics system; and
repeating the determining the performing the optimization in response to the updated model component.
10 . A system for optimizing operation of a fleet of vehicles, the system comprising:
a computer system including one or more processors in operative communication with one or more non-transitory memory media, the one or more non-transitory memory media configured with instructions executable by one or more of the one or more processors to:
provide an optimization network including a plurality of nodes and a plurality of connections between respective pairs of the plurality of nodes, the plurality of nodes defining all potential destinations in the optimization network for freight delivery or pickup, the plurality of connections defining all potential sequences of nodes traversable over the optimization network;
determine a pre-optimization simulation matrix including a plurality of sets of energy expenditures for traversing respective ones of each of the connections, each of the plurality of sets of energy expenditures accounting for permutations of a plurality of different vehicle powertrains present in a fleet of vehicles, the pre-optimization simulation matrix being configured to reduce a computational burden on the one or more processors in performing an optimization;
perform the optimization using the pre-optimization simulation matrix, fleet information indicating current locations and current energy states of a plurality of vehicles in the vehicle fleet, mission information indicating a pickup destination and a delivery destination for a plurality of items of freight, and a fleet optimization objective, wherein the optimization determines a plurality of optimized dispatch assignments each defining at least a particular vehicle of the fleet, a selection of items from the plurality of items of freight, and a sequence of nodes for delivery or pickup the selection of items;
dispatch a plurality of vehicles of the fleet of vehicles according to the plurality of optimized dispatch assignments effective to control operation of the plurality of vehicles according to the dispatching.
11 . The system of claim 10 , wherein the plurality of different vehicle powertrains comprise two or more of a combustion engine powertrain, a combustion-electric powertrain, a purely battery-electric powertrain, and a fuel-cell electric powertrain.
12 . The system of claim 10 , wherein each set of energy expenditures further accounts for one or more of:
a first plurality of points over a range of potential initial vehicle energy states;
a second plurality of points over a range of potential initial vehicle freight load states; and
a third plurality of points over a range of potential environmental conditions.
13 . The system of claim 12 , wherein the optimization performs an interpolation between two points of at least one of the first plurality of points, the second plurality of points, and the third plurality of points, the interpolation being in response to an input indicating a value intermediate said two points.
14 . The system of claim 10 , wherein each set of energy expenditures further accounts for utilization of one or more automation features in controlling one or more vehicles of the fleet.
15 . The system of claim 10 , wherein the optimization accounts for availability of one or more energy resources in determining the plurality of optimized dispatch assignments.
16 . The system of claim 15 , wherein the one or more energy resources include one or more of a diesel fueling, natural gas fueling, battery charging, battery exchange, and hydrogen filling.
17 . The system of claim 10 , wherein the fleet optimization objective comprises a minimization of net energy expenditure of the fleet of vehicles.
18 . The system of claim 10 , wherein said instructions executable by one or more of the one or more processors to:
operate a machine learning component to update a model component utilized in the determining the pre-optimization simulation matrix in response to fleet operation feedback provided by a telematics system; and
re-determine the pre-optimization simulation matrix and re-perform the optimization in response to the updated model component.