Fleet management analytics and remediation

Abstract

To provide fleet analytics and identifying a remediation action for fleet management.SOLUTION: A method and system for fleet management is described. Fleet management data for a target fleet of vehicles is received. A target vehicle vector that represents a target vehicle of the target fleet is generated based on the fleet management data. One or more similar vehicles are identified, from a plurality of reference vehicles, that are similar to the target vehicle using a distance metric between the target vehicle vector and a plurality of reference vehicle vectors. The plurality of reference vehicle vectors represent the plurality of reference vehicles. Vehicle characteristics of the target vehicle and the one or more similar vehicles are identified, the vehicle characteristics affecting vehicle efficiency of the target vehicle. The vehicle characteristics affecting vehicle efficiency of the target vehicle and corresponding vehicle characteristics of the one or more similar vehicles are displayed.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] FIELD OF THEINVENTION FIELD OF THE DISCLOSURE This application relates to fleet management, and more particularly, to methods and systems for analyzing fleet characteristics and developing modifications thereto.

[0002] Related Applications This application is related to and claims priority to U.S. Provisional Application No. 63 / 546,420, filed October 30, 2023, the disclosure of which is incorporated by reference in its entirety herein. [Background technology]

[0003] background Logistics, distribution, and delivery companies may use fleets of trucks to transport goods. Fleets may range from a few trucks to dozens or hundreds of trucks of various types and are managed by a fleet manager. Fleet managers' goals typically include reducing costs associated with operating and accessing assets (trucks, trailers, loading and unloading equipment, maintenance facilities, etc.), avoiding contingencies and risks, and maximizing asset uptime. To achieve these goals, fleet managers also seek to minimize or eliminate inefficiencies in acquiring data.

[0004] Even if data is available for an individual truck in a manager's own fleet, insights may be improved by utilizing statistical data about other trucks similar to that individual truck. For example, a fleet manager of 2019 Freightliner Class 8 trucks may know the number of miles driven, when preventive maintenance was performed, the typical fuel economy (e.g., miles per gallon) over the truck's life, etc., but maintenance and fleet management decisions may be improved by considering data from other similar trucks.

[0005] It is with respect to these and other general considerations that the embodiments have been described, and while relatively specific problems have been described, it should be understood that the embodiments should not be limited to solving the specific problems identified in the Background. Summary of the Invention [Problem to be solved by the invention]

[0006] overview Aspects of the present disclosure relate to providing fleet analytics and identifying remedial actions for fleet management. [Means for solving the problem]

[0007] In one aspect, a method for fleet management is provided. Receive fleet management data for a target fleet of vehicles. Generate a target vehicle vector representing a target vehicle of the target fleet based on the fleet management data. Identify one or more similar vehicles from a plurality of reference vehicles that are similar to the target vehicle using a distance metric between the target vehicle vector and the plurality of reference vehicle vectors. The plurality of reference vehicle vectors represent the plurality of reference vehicles. Identify vehicle characteristics of the target vehicle and the one or more similar vehicles, the vehicle characteristics impacting vehicle efficiency of the target vehicle. Display the vehicle characteristics impacting vehicle efficiency of the target vehicle and corresponding vehicle characteristics of the one or more similar vehicles.

[0008] In one embodiment, receiving the fleet management data includes receiving a maintenance history of the target fleet of vehicles.

[0009] In one embodiment, the method includes determining a remedial action for managing at least one of the subject vehicles based on a vehicle characteristic that affects vehicle efficiency of the subject vehicle.

[0010] The method may include generating a target fleet vector representing the target fleet based on the fleet management data. It may also include generating a simulated fleet vector based on the target fleet vector and a plurality of reference fleet vectors using a distance metric, where the plurality of reference fleet vectors represent a plurality of reference fleets of vehicles and the simulated fleet vector represents a simulated fleet of vehicles. Either method may include determining remedial actions for target fleet management based on the simulated fleet vector, where generating the target fleet vector includes generating, for each vehicle in the target fleet, a respective target vehicle vector representing the vehicle.

[0011] The process of generating the simulated fleet vectors may include generating, for each vehicle in the subject fleet, a simulated vehicle vector based on a corresponding subject vehicle vector and a plurality of reference vehicle vectors using a distance metric.

[0012] In one embodiment, the distance metric is the Manhattan distance metric.

[0013] In one embodiment, the process of identifying one or more similar vehicles may include using a similarity algorithm to sort a plurality of reference vehicles and a target fleet of vehicles into a plurality of groups; and calculating a distance metric between vehicles within the groups of the plurality of groups.

[0014] In one embodiment, the fleet management data may represent the workload of a target fleet of vehicles, and a simulated fleet of vehicles may be generated to perform the workload of the target fleet of vehicles.

[0015] In one embodiment, the simulated fleet of vehicles includes at least one target vehicle from a target fleet of vehicles having a modification indicated by the remedial action.

[0016] In one embodiment, the simulated fleet of vehicles has a different number of vehicles than the target fleet of vehicles, and the remedial action may indicate adding vehicles to the target fleet of vehicles or removing vehicles from the target fleet of vehicles.

[0017] In another aspect, a system for fleet management is provided that includes a processor and a non-transitory computer readable memory having computer readable instructions that, when executed by the processor, cause the processor to receive fleet management data for a subject fleet of vehicles; generate a subject vehicle vector representing a subject vehicle of the subject fleet based on the fleet management data; identify one or more similar vehicles from the plurality of reference vehicles that are similar to the subject vehicle using a distance metric between the subject vehicle vector and a plurality of reference vehicle vectors representing the plurality of reference vehicles; identify vehicle characteristics of the subject vehicle and the one or more similar vehicles that affect vehicle efficiency of the subject vehicle; and display the vehicle characteristics that affect vehicle efficiency of the subject vehicle and the corresponding vehicle characteristics of the one or more of the plurality of similar vehicles.

[0018] In some embodiments, the computer readable instructions further cause the processor to receive a maintenance history of the target fleet of vehicles.

[0019] In an embodiment, the computer readable instructions further cause the processor to determine a remedial action for management of the subject vehicle based on vehicle characteristics that affect vehicle efficiency of the subject vehicle.

[0020] The computer readable instructions may cause the processor to generate a target fleet vector representing the target fleet based on the fleet management data; generate a simulated fleet vector representing a simulated fleet of vehicles based on the target fleet vector and a plurality of reference fleet vectors representing a plurality of reference fleets of vehicles using a distance metric; and determine remedial actions for management of the target fleet based on the simulated fleet vector, where generating the target fleet vector may include generating, for each vehicle in the target fleet, a respective target vehicle vector representing that vehicle.

[0021] In some embodiments, the computer readable instructions cause the processor to generate, for each vehicle in the target fleet, a simulated vehicle vector from the multiple reference fleet vectors based on the corresponding target vehicle vector and the multiple reference vehicle vectors using a distance metric.

[0022] In some embodiments, the distance metric is a Manhattan distance metric.

[0023] In some embodiments, the computer readable instructions may further cause the processor to sort the plurality of reference vehicles and the target fleet of vehicles into a plurality of groups using a similarity algorithm; and calculate an inter-vehicle distance metric within the plurality of groups.

[0024] In some embodiments, it is contemplated that the fleet management data represents the workload of a target fleet of vehicles, and a simulated fleet of vehicles is generated to perform the workload of the target fleet of vehicles.

[0025] In certain embodiments, the simulated fleet of vehicles includes at least one target vehicle from a target fleet of vehicles having a modification indicated by the remedial action.

[0026] In some embodiments, the simulated fleet of vehicles has a different number of vehicles than the target fleet of vehicles, and the remedial action indicates adding vehicles to the target fleet of vehicles or removing vehicles from the target fleet of vehicles.

[0027] This Summary is provided to introduce some concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0028] Non-limiting and non-exhaustive examples are described with reference to the following figures. [Brief description of the drawings]

[0029] [Figure 1] FIG. 1 is a block diagram of an example system for improving fleet management according to an example embodiment. [Diagram 2] FIG. 2 is a block diagram illustrating an example of vectoring for individual tracks, according to an exemplary embodiment. [Diagram 3] FIG. 3 is an illustration of a reference vehicle identified based on a target vehicle according to an exemplary embodiment. [Figure 4] FIG. 4 is an illustration of a simulated fleet identified using the set of reference trucks identified by the fleet comparison engine of FIG. 1, according to an exemplary embodiment. [Diagram 5] FIG. 5 is an illustration of a user interface for comparing a subject fleet with multiple simulated fleets according to an illustrative embodiment. [Figure 6] FIG. 6 is an illustration of a user interface for comparing a subject fleet with multiple simulated fleets according to an illustrative embodiment. [Figure 7] FIG. 7 is an illustration of a user interface for comparing a target vehicle with multiple reference vehicles according to an exemplary embodiment. [Figure 8]FIG. 8 is a flowchart of an example method for identifying remedial actions for fleet management, according to an example embodiment. [Figure 9] FIG. 9 is a block diagram illustrating example physical components of a computing device capable of implementing aspects of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] Detailed Description In the following detailed description, reference is made to the accompanying drawings, which form a part hereof, in which specific embodiments or examples are illustrated. These aspects may be combined, other aspects may be utilized, and structural changes may be made without departing from the disclosure. The embodiments may be embodied as methods, systems, or devices. Thus, the embodiments may take the form of a hardware implementation, an entirely software implementation, or an implementation combining software and hardware aspects. Therefore, the following detailed description is not to be taken in a limiting sense, and the scope of the disclosure is defined by the appended claims and their equivalents.

[0031] As discussed in the background, maintenance and fleet management decisions can be improved by considering data from similar trucks. However, identifying similar trucks can be difficult because each truck has a wide range of characteristics, options, and usage criteria. For example, a truck may have significantly different fuel economy when it primarily operates in urban areas with many steep hills (e.g., resulting in frequent stops) where fuel economy may be poor, compared to operating long distance routes on flat ground where fuel economy may be relatively good. For example, it may be difficult to consider different usage criteria and other characteristics depending on whether a statistically significant number of trucks is available for analysis. However, trucks with sufficiently similar criteria and characteristics, even if imprecise, may be identified for useful analysis, thereby identifying remedial actions to improve fleet operations.

[0032] In various aspects, the fleet comparison engine is configured to receive fleet management data regarding a target fleet of vehicles. The target fleet of vehicles may be a fleet of trucks for a logistics company, a distribution company, a manufacturer, or other suitable fleet. To make the comparison with other reference vehicle fleets faster and more efficiently, the fleet comparison engine generates a target vehicle vector representing the target vehicles of the target fleet based on the fleet management data. In some examples, the target vehicle vector is a vector of numerical values ​​(e.g., integer values ​​and / or floating point values). The fleet comparison engine can then identify one or more similar vehicles similar to the target vehicle from the multiple reference vehicles using a distance metric between the target vehicle vector and the multiple reference vehicle vectors. The multiple reference vehicle vectors represent the multiple reference vehicles. The fleet comparison engine identifies the target vehicle and vehicle characteristics of the one or more similar vehicles that affect the vehicle efficiency of the target vehicle, and displays the vehicle characteristics that affect the vehicle efficiency of the target vehicle and the corresponding vehicle characteristics of the one or more similar vehicles. For example, the vehicle characteristics of the target vehicle may include any physical and / or driving characteristics of the target vehicle that can be measured and classified. For example, vehicle characteristics may include the subject vehicle's class, brand, model, year, height, width, length, weight, number of axles, payload, tire load rating, tire pressure, odometer reading, fuel economy average (MPG), fuel economy components (e.g., wind deflectors), mean time between failures (MTBF), center of gravity, vehicle dynamics response time, and / or driving system.

[0033] In some examples, the fleet comparison engine generates a target fleet vector representing a target fleet based on the fleet management data, and generates a simulated fleet vector based on the target fleet vector and a plurality of reference fleet vectors using a distance metric. The plurality of reference fleet vectors represent a plurality of reference vehicle fleets, and the simulated fleet vector represents a simulated fleet of vehicles. The reference fleet vector represents a plurality of reference fleets of vehicles that are different from the target fleet of vehicles, and the simulated fleet vector represents a simulated fleet of vehicles. The simulated fleet of vehicles may be considered to be a "fantasy" fleet of vehicles in that at least some individual vehicles, and possibly all vehicles in the simulated fleet, may not be operated by the fleet manager of the target fleet of vehicles. However, expanding the pool of data on which the analysis may be performed and processing the pool of data for efficient comparison improves the analysis. Thus, the fleet comparison engine may provide comparative data for understanding fleet operations and / or determine remedial actions for management of the target fleet of vehicles based on the simulated fleet vector. For example, the simulated fleet vectors may provide an indication that one or more particular trucks in the subject fleet are responsible for a larger than expected share of efficiency inefficiencies (e.g., poor miles per gallon) and could benefit from maintenance, replacement, driver training, etc.

[0034] Many other embodiments relating to computing devices are described herein. For example, FIG. 1 is a block diagram of an example system 100 for comparing and improving fleet management, according to an exemplary embodiment. The system 100 includes a fleet comparison engine 110 configured to display vehicle characteristics that affect vehicle efficiency and determine improvement actions for management of a target fleet, such as target fleet 150, based on vehicle data from different vehicles. The target fleet 150 includes a plurality of vehicles, shown as truck 152 and truck 154. Although only two trucks are shown, the target fleet 150 can have 10, 50, hundreds, or any suitable number of trucks. In some examples, all trucks in the target fleet 150 are of the same type, such as a medium truck for local delivery, a long-haul truck, or other suitable type. In other examples, the target fleet 150 has two, three, or more different types of trucks. For ease of explanation, the examples are described with respect to truck 152, but the examples are applicable to other vehicles, such as truck 154, truck 162, truck 164, or other vehicles. Although only one target fleet is shown, a business or organization may operate vehicles in two, three, or more target fleets. Additionally, while the example shown in Figure 1 uses trucks as the vehicles in target fleet 150, in other examples, target fleet 150 may include other vehicles, such as vans, cars, boats, airplanes, helicopters, etc. In one example, a first target fleet of a business includes vans, a second target fleet includes trucks, and a third target fleet includes airplanes.

[0035] Truck 152 may communicate with target fleet data store 120 to provide vehicle data for storage, analysis, and fleet management, for example. Truck 152 vehicle data may include truck class, truck brand, truck model, year, diagnostic data, odometer reading, fuel consumption, fuel economy average (MPG), mean time between failures (MTBF), vehicle telematics data, location data (e.g., GPS coordinates), driver management (e.g., attention and driving characteristics), or other suitable data for fleet management applicable to a particular vehicle. That is, truck 152 vehicle data may include physical characteristics and / or driving characteristics associated with truck 152. In some examples, vehicle data is captured by one or more sensors and processors (not shown) within truck 152 used in normal operation of truck 152, such as fuel sensors, engine sensors (e.g., fuel pressure, oil temperature, RPM, etc.), engine controller, GPS receiver, dashboard or informatics / infotainment display controller, etc. In other examples, truck 152 includes a stand-alone processor or device (e.g., a GPS receiver) that captures some of the vehicle data. Truck 152 can include communication device 153 or other suitable telematics device, which can include a mobile network interface (e.g., a 3G / 4G / 5G transceiver), a satellite interface, a Wi-Fi interface, a Bluetooth interface, a data port, or other suitable device for transmission of vehicle data. Communication device 153 can obtain vehicle data from sensors and / or processors (e.g., via a CAN bus, an OBD-III port, or other suitable port). Transmission of vehicle data can utilize network 180, which can include one or more networks, such as a local area network (LAN), a wide area network (WAN), an enterprise network, the Internet, etc., and can include one or more wired, wireless, and / or optical portions.In some examples, the communication device 153 may communicate with a smartphone, tablet, laptop, etc. as an intermediary to the network 180 or the target fleet data store 120.

[0036] The target fleet data store 120 may be a computing device, a network server, a cloud storage service, a database, or any other suitable data store. In general, the target fleet data store 120 stores fleet information regarding the target fleet 150. In some examples, the target fleet data store 120 processes the fleet information and also provides a user interface for software with fleet management functions (e.g., route planning, maintenance planning, fleet status). The fleet information may preferably include vehicle data from each of the vehicles in the target fleet 150 along with additional data related to fleet management (e.g., owner information, fleet-related service contracts, maintenance contracts, trailer information, cargo details (including weight and capacity utilization), driver profiles, spare parts inventory information, route planning information, etc.). The fleet information may also include location data received from a GPS service, fuel consumption data received from a gas station, tractor-trailer information received from a weigh station, weather, road conditions, etc.

[0037] The system 100 may also include one or more maintenance facilities, such as a maintenance facility 140. The maintenance facility 140 may be a truck service station, a fuel station, a repair station, or other suitable facility for maintenance of the truck 152. In some examples, the maintenance facility 140 includes a diagnostic device (not shown) that receives vehicle data regarding the truck 152. As an example, the diagnostic device is a computing device or tablet with a user interface for inputting data. For example, a technician performing an oil change on the truck 152 may input information regarding the oil change, such as the current mileage, the oil viscosity of the new oil, and the oil analysis of the used oil, into the user interface. In another example, the diagnostic device is an engine diagnostic device that communicates with the truck 152 via, for example, an OBD-III port, a 9-pin type 2 port, a controller area network (CAN) bus port, or other suitable interface.

[0038] As described above, the fleet comparison engine 110 is configured to determine remedial actions for the management of the target fleet 150. To do so, the fleet comparison engine 110 utilizes vehicle data from the target fleet 150 along with vehicle data from a plurality of reference fleets 160. Each reference fleet 160 may correspond to a reference fleet data store 170, which generally corresponds to the target fleet 150 and the target fleet data store 120, respectively. For ease of explanation, a single instance of the reference fleet 160 and the reference fleet data store 170 is described in various examples, although additional instances (e.g., 50, 100, 300 or more instances) may be used in other examples. The reference fleet 160 may include additional trucks (not shown) having different levels of similarity to the trucks of the target fleet 150.

[0039] Reference fleet 160 includes trucks 162 and 164, which may be similar to trucks 152 and 154 of target fleet 150. That is, trucks 162 and 164 may each have a communication device (not shown) suitable for providing vehicle data to reference fleet data store 170. Additionally, trucks 162 and 164 may visit maintenance facility 140 for maintenance. The vehicle data for trucks 162 and 164 may be provided to fleet comparison engine 110 via reference fleet data store 170, directly from trucks 162 and 164, or in any suitable combination.

[0040] FIG. 2 illustrates a block diagram showing an example of vectorization for individual trucks, according to an exemplary embodiment. As described above, the fleet comparison engine 110 is configured to identify vehicle characteristics of the target vehicle that affect the vehicle efficiency of the target vehicle, and can determine remedial actions for management of the target fleet 150 based on vehicle data from different vehicles. The different vehicles can include vehicles in the target fleet 150 and vehicles in the reference fleet 160. In some situations, the fleet comparison engine 110 can have access to thousands or tens of thousands of vehicle data from multiple different fleets. To improve processing speed and efficiency for determining suitable remedial actions, the fleet comparison engine 110 generates a target fleet vector 200 representing the target fleet 150 and a similar reference fleet vector (not shown) for the reference fleet 160.

[0041] The target fleet vector 200 is based on vehicle data 252 for truck 152, vehicle data 254 for truck 154, etc. In some examples, the fleet comparison engine 110 generates a target vehicle vector for each truck, such as a target vehicle vector 253 for truck 152 and a target vehicle vector 255 for truck 154. In the example shown in FIG. 2, the vehicle data 252 includes truck class, truck brand, truck model, year, odometer reading, average fuel economy (MPG), service area, and mean time between failures (MTBF). In other examples, additional criteria may be included in or omitted from the vehicle data 252. Similar information for truck 154 is displayed as vehicle data 254. In the illustrated example, the target fleet vector 200 is generated by concatenating the target vehicle vectors (e.g., 253, 355) of the target fleet 150.

[0042] In the example shown in FIG. 2, the target vehicle vector 253 is based on the vehicle data 252 values, where some values ​​are copied as integers (e.g., class and year) and other values ​​are converted from text to integer indexes (e.g., converting Freightliner to 0, Volvo to 1, etc.). While the example of FIG. 2 shows only integer values, in other examples, floating point values ​​may be used in addition to or instead of integer values. In some examples, one or more values ​​from the vehicle data 252 may be processed to obtain a single value for the target vehicle vector 253. For example, the fleet comparison engine 110 may perform a hash function on the class, odometer, and service area to obtain a hybrid value for the target vehicle vector 253. The hybrid value may be generated instead of or in addition to the value on which the hybrid value is based and inserted into the target vehicle vector 253. In some examples, two, three, or more hybrid values ​​are used to improve matchmaking between vehicles. In still other examples, some values ​​may be weighted to have a higher impact on the identified remedial action.

[0043] The hybrid value may also represent a parameter based on other values ​​of the vehicle data. For example, the utilization of a particular truck may not be available as a value directly from the vehicle data, but may be determined based on, for example, a location map of where the truck is located. In one example, utilization is the percentage of time the truck is performing useful tasks, such as loading, unloading, or driving, as opposed to idling at a rest stop, undergoing maintenance, or parked. Utilization may be determined based on geolocation data to map where the truck has traveled. In one example, a truck may be considered utilized when the Manhattan or Euclidean distance of the GPS location is greater than 2 miles, 5 miles, or other suitable threshold. In another example, a truck may be considered utilized when it is located outside of a predefined geofence area.

[0044] In some examples, the fleet comparison engine 110 is configured to group the vehicles based on a particular characteristic of interest or using a similarity algorithm. For example, the fleet comparison engine 110 can perform locality sensitive hashing or other suitable similarity algorithms to sort the vehicles into groups. In general, the similarity algorithm is configured to identify groups, nearest neighbors, or subsets of vehicles from the target fleet 150 and / or the reference fleet 160, where members of the groups are similar in one or more criteria or characteristics. For example, a first group can include trucks with payloads between 20,000 pounds and 30,000 pounds, and a second group can include trucks with payloads between 30,000 pounds and 60,000 pounds. Other suitable similarity algorithms include the MinHash algorithm, tree-based techniques such as regression trees and classification trees, and the like. In the case of LSH (locality sensitive hashing) as the similarity algorithm, LSH can be used to indicate that similar vehicles belong to the same hash group or "bucket," while dissimilar vehicles belong to different groups. In various examples, different hashes can be used to represent groups of vehicles separated by type, class, region, or a combination thereof (e.g., Class 8 sleeper trucks located in the Midwest). Advantageously, the fleet comparison engine 110 can perform an approximate nearest neighbor search using only one vehicle in a hash group, which in some examples can take as little as O(n 2 ) to O(n). For example, the fleet comparison engine 110 can calculate distance metrics between vehicles in a group instead of between all available reference vehicles.

[0045] In some examples, the target fleet vector includes aggregate information based on vehicle data (e.g., by aggregating or concatenating target vehicle vectors for target fleets). For example, the target fleet vector may indicate the number of vehicles in the Midwest region, the number of vehicles in the Southeast region, the number of over-the-road vehicles, the number of Class 8 vehicles, etc.

[0046] In addition to generating the target fleet vector, the fleet comparison engine 110 generates a reference fleet vector that corresponds to at least a portion of the reference fleet 160. The reference fleet vector is generated similarly to the target fleet vector, but is based on the vehicle data of the reference fleet. The vehicle fleet comparison engine 110 can then use the vectors (i.e., the target fleet vector and the reference fleet vector) to compare the target fleet of vehicles to the reference fleet of vehicles, instead of comparing individual trucks. Advantageously, the target fleet vector and the reference fleet vector can include hybrid values ​​that emphasize desirable characteristics, such as fuel economy, over less desirable characteristics (e.g., truck brand). As an example, a fleet manager of a target vehicle with a particular vehicle configuration (e.g., 60% Class 8 trucks, 40% Class 7 trucks) can compare the target vehicle to other similar reference vehicles with the same or similar configuration within the same industry.

[0047] A comparison of the subject fleet to the reference fleet can then be performed using each of the values ​​in the vector, or only a subset of the values, in various examples. In some situations, a comparison based on one or two hybrid values ​​is processed more quickly than a comparison based on all values ​​in the vector. In other situations, a comparison of additional values ​​in the vector may be used to identify a more similar reference fleet.

[0048] Although target fleet vector 200 is shown as a one-dimensional vector or array, target fleet vector 200 may in other examples be implemented as a multi-dimensional vector, a multi-dimensional matrix, or other suitable data structure. Additionally, target fleet vector 200 may in some examples be maintained as separate vectors, such as a separate vector for each vehicle in the target fleet, or a separate vector for each class of vehicle, each region, etc.

[0049] The fleet comparison engine 110 allows fleet managers to evaluate the relative performance of similar trucks, identify outliers, and identify potential causes of outliers with poor performance. By using the target fleet vector and the reference fleet vector along with one or more distance metrics, the fleet comparison engine 110 efficiently identifies which reference fleets and, in turn, which trucks are suitable for comparison. In some examples, the distance metric is a cosine similarity metric that is performed using the values ​​of the target fleet vector and each of the reference fleet vectors to obtain a similarity score of the target fleet vector to the corresponding reference fleet vector. In this way, a similarity value of the target fleet to the reference fleet is obtained, and reference fleets with a similarity value above a threshold (e.g., greater than 0.9) or within the "top 5" of similarity, for example, can be selected for further analysis.

[0050] FIG. 3 illustrates a diagram of reference vehicles identified based on a target vehicle, such as truck 152. As described above, the fleet comparison engine 110 can be configured to identify a simulated fleet of vehicles based on the target fleet and the reference fleet. In the example illustrated in FIG. 3, to assemble the simulated fleet, the fleet comparison engine 110 uses a suitable distance metric (described above) to compare the target vehicle vector for truck 152 with the reference vehicle vectors of other trucks in the reference fleet data store 170. The fleet comparison engine 110 can generate a similarity score for each pairing of trucks (i.e., one truck from the reference fleet paired with truck 152) and select a set of trucks to compose the simulated fleet, as described below. In the example illustrated in FIG. 3, the set of trucks includes the five reference trucks (shown as trucks 310, 312, 314, 316, 318) with the highest similarity scores to the target truck 152. The set of trucks can be ranked according to one or more suitable criteria. In the example shown in Figure 3, trucks 310, 312, 314, 316, 318 are ranked according to fuel economy (MPG). In other examples, trucks are ranked according to utilization rate, mean time between failures, carbon dioxide emissions, cost of ownership, fleet availability, or other suitable parameters. The fleet comparison engine 110 can continue with the above process to identify a different set of reference trucks for each truck in the target fleet. Thus, for a target fleet having five trucks, the fleet comparison engine 110 can generate five sets of five reference trucks.

[0051] As noted above, the reference fleet data store 170 may include vehicle data for trucks other than the target fleet, and may even be from competing fleets. In some examples, portions of the vectors may be anonymized to prevent personal or proprietary information from being disclosed to fleet managers of the target fleet.

[0052] FIG. 4 shows a diagram of simulated fleets identified using a set of reference trucks identified by the fleet comparison engine 110. The simulated fleets of vehicles may be considered "fantasy" fleets of vehicles in that at least some individual vehicles may not be operated by the fleet manager of the subject fleet of vehicles. Using the example of FIG. 3, five simulated fleets are identified, each having five trucks. Each truck is identified by a fleet number and a truck number, so that the first simulated fleet 410 includes trucks 1-1, 2-1, 3-1, 4-1, and 5-1, the third simulated fleet 430 includes trucks 1-3, 2-3, 3-3, 4-3, and 5-3, and the fifth simulated fleet 450 includes trucks 1-5, 2-5, 3-5, 4-5, and 5-5. For clarity, the second and fourth simulated fleets are not shown. Each of the simulated fleets includes one truck that corresponds to one truck in the subject fleet, and the trucks are arranged among the simulated fleets such that similar vehicles that are all the same ranked are grouped into the same simulated fleet. That is, simulated fleet 410 is the "best" simulated fleet, including reference truck 310 and a similarly ranked (i.e., number 1) reference truck from the other set of reference trucks. Simulated fleet 450 is the "worst" simulated fleet, including reference truck 318 and a similarly ranked (i.e., number 5) reference truck from the other set of reference trucks.

[0053] 5, 6, 7, 8, and 9 show examples of user interfaces that can be generated by the fleet comparison engine 110, or suitable data for the user interface 500 can be provided for rendering by a client computing device.

[0054] 5 shows a diagram of a user interface 500 for comparing a target fleet to multiple simulated fleets. In FIG. 5, the average fuel economy across a fleet of trucks is compared between the target fleet and ten simulated fleets (e.g., similar to simulated fleets 410, 430, 450). Thus, the fleet comparison engine 110 identifies the relative efficiency of the target fleet for a fleet manager, which can be used to identify remedial actions for the target fleet.

[0055] Figure 6 shows a diagram of a user interface 600 for comparing a subject fleet to multiple simulated fleets. In Figure 6, the standard deviation of fuel economy (MPG) is plotted against the fleet average fuel economy, which indicates that the subject fleet's operations can be improved by making fuel consumption more consistent (e.g., improving budgeting capabilities) and improving average fuel economy (reducing operating costs).

[0056] FIG. 7 shows a diagram of a user interface 700 for comparing a target vehicle (target truck 702) with multiple reference vehicles (reference truck 1, reference truck 10). In FIG. 7, the user interface 700 provides a visual comparison of a target vehicle from a target fleet with similar reference trucks from a simulated fleet. In this example, reference truck 10 (column 710) corresponds to the worst performing truck (in terms of fuel economy) among trucks similar to the target truck 702, while reference truck 1 (column 718) corresponds to the best performing truck (in terms of fuel economy) among the similar trucks. As shown in FIG. 7, the fleet comparison engine 110 can highlight values ​​from the corresponding vector that are far from the target truck 702 according to a preferred distance metric, as described above. In the example shown in FIG. 7, the fleet comparison engine 110 highlights the idle time percentage (Idle Hours Pct), which indicates that the worst performing truck (710) is idle 28% of the time, while the best performing truck (718) is idle only 11% of the time. In this example, the fleet comparison engine 110 may determine remedial actions for management of the subject fleet that includes the subject truck 702, where the remedial actions may include driver training to reduce idling, rerouting delivery routes to avoid inefficient truck stops, adding auxiliary power units (APUs) to provide trailer refrigeration without running the truck's motor, or other suitable remedial actions. The fleet comparison engine 110 may also highlight different transmission types and determine remedial actions including changing the transmission from a manual transmission to an automatic transmission, changing to a truck with an automatic transmission, etc.

[0057] FIG. 8 illustrates a flow chart of an exemplary method 800 for identifying remedial actions for fleet management, according to an exemplary embodiment. The technical processes illustrated in these figures are performed automatically unless otherwise noted. In any embodiment, some steps of the process can be repeated, possibly operating with different parameters or data. Steps in an embodiment may be performed in a different order from the top-down order listed in FIG. 8. Steps may be performed sequentially, partially overlapping, or fully in parallel. Thus, the order in which steps of method 800 are performed may vary from one execution of the process to another. Also, steps may be omitted, combined, renamed, grouped, performed on one or more machines, or otherwise deviated from the illustrated flow, provided that the process performed is operable and complies with at least one claim. The steps of FIG. 8 may be performed by the fleet comparison engine 110 or other suitable computing device.

[0058] Method 800 begins at step 802. In step 802, fleet management data (e.g., vehicle data, fleet information) for a subject fleet of vehicles is received. For example, fleet comparison engine 110 can receive vehicle data from subject fleet 150 (e.g., trucks 152 or 154), fleet information from subject fleet data store 120, fleet information from reference fleet data store 170, and / or vehicle data from reference fleet 160. In some examples, step 802 includes receiving a maintenance history of the subject fleet of vehicles, for example, from maintenance facility 140. In some examples, the vehicle management data represents a workload of the subject fleet of vehicles, and a simulated fleet of vehicles is generated to perform the workload of the subject fleet of vehicles.

[0059] In step 804, a target vehicle vector is generated that represents target vehicles in a target fleet based on the fleet management data.

[0060] In step 806, one or more similar vehicles are identified from the plurality of reference vehicles that are similar to the target vehicle using a distance metric between the target vehicle vector and the plurality of reference vehicle vectors. The plurality of reference vehicle vectors represent a plurality of reference vehicles. In some examples, the distance metric is a Euclidean distance metric, a Manhattan distance metric, or a Jaccard similarity metric. For example, the plurality of reference vehicles and the target fleet of vehicles are sorted into a plurality of groups using a locality-sensitive hashing process (or other suitable similarity algorithm as described above). A suitable distance metric is then used to calculate the distance between vehicles within a particular one of the plurality of groups. The distance metric is calculated only within the group, and not for all of the reference and target vehicles, thereby reducing processing time and resources.

[0061] In step 808, vehicle characteristics of the subject vehicle and one or more similar vehicles are identified. Specifically, vehicle characteristics of the subject vehicle that affect the vehicle efficiency of the subject vehicle are identified. As described above, vehicle characteristics may include physical characteristics and / or driving characteristics associated with the subject vehicle. For example, once one or more similar vehicles are identified, the vehicle characteristics of the subject vehicle and the one or more similar vehicles are compared based on the vehicle management data to evaluate relative performance. Based on the comparison, one or more vehicle characteristics are identified as affecting the vehicle efficiency of the subject vehicle. For example, changes to certain physical characteristics or driving characteristics that may be useful improve or change the vehicle efficiency of the subject vehicle.

[0062] It should be appreciated that in some embodiments, a user may provide or select one or more vehicle characteristics of the subject vehicle to evaluate its relative performance compared to one or more similar vehicles. For example, a user may be specifically interested in the fuel efficiency of a subject truck, and may compare the fuel consumption rate of the subject vehicle to one or more similar vehicles.

[0063] In step 810, vehicle characteristics that affect vehicle efficiency of the subject vehicle and corresponding vehicle characteristics of one or more similar vehicles are displayed.

[0064] In some examples, the method 800 further includes determining remedial actions for management of the subject vehicle based on vehicle characteristics that affect the vehicle efficiency of the subject vehicle. Different remedial actions may be determined by the fleet comparison engine 110. Examples of remedial actions include notification of deficiencies or delays in preventive maintenance routines to reduce mean time between failures and / or mean time to recovery, recommending an aerodynamic package (e.g., a body kit) to improve fuel economy, installing an auxiliary power unit to reduce idling time, changing driver routes to avoid long stops at inefficient truck stops, recommending driver training to reduce idling time, recommending further analysis of the vehicle for trade-in consideration, recommending switching to a vehicle with an automatic transmission to improve fuel economy, recommending switching to a vehicle with a different engine to improve fuel economy, recommending tire pressure adjustment to improve tire wear, etc.

[0065] The method 800 may further include generating a target fleet vector representing the target fleet based on the fleet management data; generating a simulated fleet vector based on the target fleet vector and a plurality of reference fleet vectors using a distance metric; and determining remedial actions for management of the target fleet based on the simulated fleet vector. The plurality of reference fleet vectors represent a plurality of reference fleets of vehicles, and the simulated fleet vector represents a simulated fleet of vehicles. Generating the target fleet vector may include generating, for each vehicle in the target fleet, a respective target vehicle vector representing the vehicle.

[0066] The target fleet vector may correspond to the target fleet vector 200. Thus, the target fleet vector may include one or more target vehicle vectors 253, 255, etc. In some examples, the target fleet vector includes additional fleet information related to, for example, route plans, maintenance plans, fleet status, etc. Additionally, the simulated fleet vector may generally correspond to the target fleet vector, but may use vehicle data from similar vehicles, as discussed above with respect to Figures 3-5. For example, the simulated fleet vector may correspond to simulated fleet 410, simulated fleet 430, or other suitable simulated fleet.

[0067] In some examples, the vehicle management data represents a workload of a target fleet of vehicles, and a simulated fleet of vehicles is generated to perform the workload of the target fleet of vehicles. Additionally, the simulated fleet of vehicles can include at least one target vehicle from the target fleet of vehicles having a modification indicated by the remedial action. In some examples, the simulated fleet of vehicles has a different number of vehicles than the target fleet of vehicles, and the remedial action indicates adding or removing a vehicle from the target fleet of vehicles.

[0068] In some examples, generating the simulated fleet vector includes, for each vehicle in the target fleet, generating a simulated vehicle vector based on a corresponding target vehicle vector and a plurality of reference vehicle vectors using a distance metric.

[0069] In some examples, using the simulated fleet, the fleet comparison engine 110 can generate estimated or predicted operating parameters based on changes in the simulated fleet relative to the subject fleet. In one example, the fleet comparison engine 110 can replace subject vehicles in the subject fleet with newer vehicles and estimate operating parameters such as fuel efficiency (fleet miles per gallon), utilization rate, mean time between failures, etc. Other changes can include changing the engine horsepower of the vehicles, adding an aero package to the vehicles, adjusting driver behavior (reducing idling), adjusting route planning, changing maintenance frequency, etc.

[0070] Figure 9 and the associated discussion provide a discussion of various operating environments in which aspects of the present disclosure may be implemented. However, the devices and systems illustrated and discussed with respect to Figure 9 are for purposes of example and explanation, and are not intended to limit the vast number of computing device configurations that may be utilized to implement aspects of the present disclosure, as described herein.

[0071] FIG. 9 is a block diagram illustrating the physical components (e.g., hardware) of a computing device 900 on which aspects of the disclosure can be implemented. Although one computing device is shown, it should be understood that the present invention can be implemented on multiple computing devices. The computing device components described below can have computer-executable instructions for implementing a fleet management application 920 on a computing device (e.g., fleet comparison engine 110), including computer-executable instructions for a fleet management application 920 executable to implement the methods disclosed herein. In a basic configuration, the computing device 900 can include at least one processing unit 902 and a system memory 904. Depending on the configuration and type of computing device, the system memory 904 can include, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read-only memory), flash memory, or any combination of such memory. The system memory 904 may include an operating system 905 and one or more program modules 906 suitable for executing a fleet management application 920, such as one or more components related to Figures 1 and 3, in particular a fleet comparison engine 921 (e.g., corresponding to fleet comparison engine 110).

[0072] The operating system 905 may be suitable, for example, for controlling the operation of the computing device 900. Furthermore, embodiments of the present disclosure may be implemented in combination with a graphics library, other operating systems, or any other application programs, and are not limited to any particular application or system. This basic configuration is illustrated in FIG. 9 by those components within dashed line 908. The computing device 900 may have additional features or functionality. For example, the computing device 900 may also include additional data storage devices (removable and / or non-removable), such as, for example, magnetic disks, optical disks, or tapes. Such additional storage devices are illustrated in FIG. 9 by removable storage device 909 and non-removable storage device 910.

[0073] As mentioned above, a number of program modules and data files can be stored in the system memory 904. While executing on the processing unit 902, the program modules 906 (e.g., fleet management application 920) can perform processes including, but not limited to, aspects described herein. Other program modules that can be used to identify remedial actions for fleet management, among others, in accordance with aspects of the present disclosure can include a fleet comparison engine 921.

[0074] Additionally, embodiments of the present disclosure may be implemented on electrical circuits including discrete electronic elements, packaged or integrated electronic chips including logic gates, circuits utilizing a microprocessor, or a single chip including electronic elements or a microprocessor. For example, embodiments of the present disclosure may be implemented via a system-on-chip (SOC) in which each or many of the components shown in FIG. 9 may be integrated onto a single integrated circuit. Such an SOC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all integrated (or "burned") onto a chip substrate as a single integrated circuit. When operating via a SOC, the functionality described herein with respect to a client being able to switch protocols may operate via application-specific logic integrated with other components of the computing device 900 on a single integrated circuit (chip). Embodiments of the present disclosure may also be implemented using other technologies capable of performing logical operations, such as, for example, AND, OR, and NOT, including, but not limited to, mechanical, optical, fluidic, and quantum technologies. Additionally, embodiments of the present disclosure may be implemented within a general-purpose computer or other circuits or systems.

[0075] The computing device 900 may also have one or more input devices 912, such as a keyboard, mouse, pen, sound or voice input device, touch or swipe input device, etc. It may also include output devices 914, such as a display, speakers, printer, etc. The above devices are by way of example only and others may be used. The computing device 900 may include one or more communication connections 916 that allow communication with other computing devices 950. Examples of suitable communication connections 916 include, but are not limited to, radio frequency (RF) transmitter, receiver, and / or transceiver circuitry, a universal serial bus (USB), a parallel port, and / or a serial port.

[0076] As used herein, the term computer readable media can encompass computer storage media. Computer storage media can include volatile and non-volatile, removable and non-removable media implemented in any manner or technology for storing information, such as computer readable instructions, data structures, program modules, etc. System memory 904, removable storage device 909, and non-removable storage device 910 are all examples of computer storage media (e.g., memory storage). Computer storage media can include RAM, ROM, Electrically Erasable Read Only Memory (EEPROM), Flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical storage devices, magnetic cassettes, magnetic tapes, magnetic disk storage devices or other magnetic storage devices, or any other article of manufacture that can be used to store information and that can be accessed by computer device 900. Such computer storage media may be part of computer device 900. Computer storage media does not include carrier waves or other propagated or modulated data signals.

[0077] Communication media may be embodied by computer readable instructions, data structures, program modules or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and includes any information delivery media. The term "modulated data signal" may refer to a signal that has one or more characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared and other wireless media.

[0078] The description and illustration of one or more aspects provided herein are not intended to limit or restrict in any way the scope of the disclosure as set forth in the claims. The aspects, examples, and detailed description provided herein are believed to be sufficient to transfer ownership and enable others to make and use the best aspects of the claimed disclosure. The claimed disclosure should not be construed as being limited to any aspect, example, or detailed description provided herein. Various features (both structures and methods), whether shown and described in combination or shown and described separately, are intended to be selectively included or omitted to produce an embodiment having a particular set of features. Although the description and illustration of the present application have been provided, those skilled in the art may envision variations, modifications, and alternative aspects that fall within the spirit of the broader aspects of the general inventive concepts embodied herein without departing from the broader scope of the claimed disclosure.

[0079] The use of any and all examples or exemplary language (e.g., "such as") provided herein is intended merely to better illuminate the disclosed embodiments and does not limit the scope of the disclosed embodiments unless otherwise stated. Numerous modifications and adaptations will be readily apparent to those of ordinary skill in the art. [Explanation of symbols]

[0080] 100 Systems 110 Fleet Comparison Engine 120 Target Fleet Datastores 140 Maintenance Equipment 150 Target Fleet 152 Tracks 153 Communication Devices 154 Tracks 160 Baseline Fleet 162 Tracks 164 Tracks 170 Baseline Fleet Datastore 180 Network

Claims

1. 1. A method for fleet management, comprising: receiving fleet management data for a target fleet of vehicles; generating a target vehicle vector representing target vehicles of the target fleet based on the fleet management data; identifying one or more similar vehicles from a plurality of reference vehicles that are similar to the target vehicle using a distance metric between the target vehicle vector and a plurality of reference vehicle vectors representing a plurality of reference vehicles; identifying vehicle characteristics of the subject vehicle and the one or more similar vehicles that affect vehicle efficiency of the subject vehicle; and Displaying vehicle characteristics that affect vehicle efficiency of the subject vehicle and corresponding vehicle characteristics of the one or more similar vehicles. The method includes:

2. The method of claim 1 , wherein the step of receiving fleet management data includes receiving a maintenance history for the target fleet of vehicles.

3. The method of claim 1 , further comprising determining remedial actions for management of the subject vehicle based on the vehicle characteristics that affect vehicle efficiency of the subject vehicle.

4. generating a target vehicle vector representing the target fleet based on the fleet management data; generating a simulated fleet vector representing a simulated fleet of vehicles based on the subject fleet vector and a plurality of reference fleet vectors representing a plurality of reference fleets of vehicles using the distance metric; and determining remedial actions for management of the target fleet based on the simulated fleet vectors; Further comprising: The method of claim 3 , wherein generating the target fleet vectors comprises: generating, for each vehicle in the target fleet, a respective target vehicle vector representing the vehicle.

5. 5. The method of claim 4, wherein generating the simulated fleet vector comprises: for each vehicle in the target fleet, using the distance metric to generate a simulated vehicle vector based on a corresponding target vehicle vector and the plurality of reference vehicle vectors.

6. The method of claim 4 , wherein the distance metric is a Manhattan distance metric.

7. The step of identifying one or more similar vehicles further comprises: sorting the plurality of reference vehicles and the target fleet of vehicles into a plurality of groups using a similarity algorithm; and calculating a distance metric between vehicles within said plurality of groups; The method of claim 4 , comprising:

8. the fleet management data represents the workload of the target fleet of vehicles; and the simulated fleet of vehicles is generated to perform the workload of the target fleet of vehicles; The method according to claim 4.

9. The method of claim 8 , wherein the simulated fleet of vehicles includes at least one target vehicle from the target fleet of vehicles having a modification indicated by the corrective action.

10. The simulated fleet of vehicles has a different number of vehicles than the target fleet of vehicles; and the remedial action directs the addition of a vehicle to the target fleet of vehicles or the removal of a vehicle from the target fleet of vehicles; The method according to claim 8.

11. 1. A system for fleet management, comprising: A processor; a non-transitory computer readable memory having computer readable instructions that, when executed by the processor, cause the processor to: receiving fleet management data for a target fleet of vehicles; generating a target vehicle vector representing target vehicles of the target fleet based on the fleet management data; identifying one or more similar vehicles from a plurality of reference vehicles that are similar to the target vehicle using a distance metric between the target vehicle vector and a plurality of reference vehicle vectors representing a plurality of reference vehicles; Identifying vehicle characteristics of the subject vehicle and the one or more similar vehicles that affect vehicle efficiency of the subject vehicle; and a non-transitory computer readable memory that displays the vehicle characteristics that affect vehicle efficiency of the subject vehicle and corresponding vehicle characteristics of the one or more similar vehicles. A system that provides.

12. The computer readable instructions further include causing the processor to: receiving a maintenance history for the target fleet of vehicles; The system of claim 11.

13. The computer readable instructions further include causing the processor to: The system of claim 11 , further comprising determining remedial actions for management of the subject vehicle based on the vehicle characteristics that affect vehicle efficiency of the subject vehicle.

14. The computer readable instructions further include causing the processor to: generating a subject fleet vector representing the subject fleet based on the fleet management data; generating a simulated fleet vector representing a simulated fleet of vehicles based on the subject fleet vector and a plurality of reference fleet vectors representing a plurality of reference fleets of vehicles using the distance metric; and determining remedial measures for management of the subject fleet based on the simulated fleet vector; generating the target fleet vectors includes, for each vehicle in the target fleet, generating a respective target vehicle vector representing the vehicle. The system of claim 13.

15. The computer readable instructions further include causing the processor to:

15. The system of claim 14, further comprising: for each vehicle in the target fleet, generating a simulated vehicle vector from the plurality of reference fleet vectors using a distance metric based on a corresponding target vehicle vector and the plurality of reference vehicle vectors.

16. The system of claim 15 , wherein the distance metric is a Manhattan distance metric.

17. The computer readable instructions further include causing the processor to: sorting the plurality of reference vehicles and the target fleet of vehicles into a plurality of groups using a similarity algorithm; and calculating a distance metric between vehicles within said plurality of groups; The system of claim 15.

18. the fleet management data represents the workload of the target fleet of vehicles; and the simulated fleet of vehicles is generated to perform the workload of the target fleet of vehicles; The system of claim 14.

19. 20. The system of claim 18, wherein the simulated fleet of vehicles includes at least one target vehicle from the target fleet of vehicles having a modification indicated by the remedial action.

20. The simulated fleet of vehicles has a different number of vehicles than the target fleet of vehicles; and the remedial action directs the addition of a vehicle to the target fleet of vehicles or the removal of a vehicle from the target fleet of vehicles; 20. The system of claim 18.

Images