Fuel delivery system

Abstract

To provide a fuel delivery system capable of efficiently delivering fuel to a plurality of users with different consumption states including the number of operating industrial vehicles and an operation time, and the like.SOLUTION: A fuel delivery system acquires a consumption state including a capacity of a fuel tank facility, a residual quantity of the tank facility, the number of operating industrial vehicles, an industrial vehicle operation time, and a fuel consumption of the industrial vehicles about each of a plurality of users, and generates a delivery plan by calculation using machine learning on the basis of the consumption state and the capacity of the tank facility. The fuel delivery system evaluates the consumption speeds of the plurality of users on the basis of the consumption state and the capacity of the tank facility, classifies the plurality of users into a first group for regularly delivering fuel in a prescribed first cycle and a second group for irregularly delivering fuel in a second cycle longer than the first cycle, and tunes delivery timing for a second group user to delivery timing for a first group user.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present invention relates to a fuel delivery system. [Background technology]

[0002] Conventionally, a technology related to a fuel delivery system is known that estimates the time to refuel a construction machine based on operation information of the construction machine and information on the remaining fuel, and creates a delivery plan for delivering fuel based on the estimated refueling time and position information of the construction machine (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-112799 A Summary of the Invention [Problem to be solved by the invention]

[0004] In general, the consumption status including the number of operating industrial vehicles and operating time of the industrial vehicles does not match among multiple users who use fuel-operated industrial vehicles. Therefore, the delivery timing generated based on the remaining fuel amount does not necessarily match, and the number of deliveries to deliver fuel to multiple users may be increased excessively. [Means for solving the problem]

[0005] One aspect of the present invention is a fuel delivery system that generates a delivery plan for delivering fuel to multiple users who use industrial vehicles powered by fuel, and includes a consumption status acquisition unit that acquires consumption status for each of the multiple users, including the capacity of the fuel tank equipment, the remaining amount in the tank equipment, the number of industrial vehicles in operation, the operating time of the industrial vehicles, and the fuel consumption of the industrial vehicles, and a delivery plan generation unit that generates a delivery plan based on the consumption status and the capacity of the tank equipment through calculations using machine learning to optimize the delivery order, required time, and transportation costs for the multiple users.The delivery plan generation unit evaluates the consumption rates of the multiple users based on the consumption status and the capacity of the tank equipment, and based on the evaluation results of the consumption rates, classifies the multiple users into a first group to which fuel is delivered periodically at a predetermined first period and a second group to which fuel is delivered irregularly at a second period longer than the first period, and generates a delivery plan through calculations using machine learning that match the timing of delivery to users in the second group to the regular timing of delivery to users in the first group.

[0006] According to a fuel delivery system of an embodiment of the present invention, the consumption rates of a plurality of users are evaluated based on the consumption status and the capacity of the tank equipment. Based on the evaluation result of the consumption rate, the plurality of users are classified into a first group to which fuel is delivered periodically at a predetermined first cycle, and a second group to which fuel is delivered irregularly at a second cycle longer than the first cycle. A delivery plan is generated by a calculation using machine learning that matches the delivery timing to the second group of users with the regular delivery timing to the first group of users. As a result, the number of deliveries can be reduced and the increase in delivery costs can be suppressed compared to a delivery plan in which, for example, delivery to the first group of users and delivery to the second group of users are performed separately. Therefore, according to the fuel delivery system of an embodiment of the present invention, it is possible to efficiently deliver fuel to a plurality of users having different consumption statuses, including the number of operating industrial vehicles and the operating time.

[0007] In one embodiment, the fuel is a biomass fuel that is discarded after the expiration date, and the delivery plan generating unit further classifies users of the second group who are estimated to have a second period that exceeds the expiration date when the next delivery amount is set to the first delivery amount based on the evaluation result of the consumption rate into a third group, and sets the next delivery amount of the users of the third group to a second delivery amount that is smaller than the first delivery amount and does not exceed the expiration date in the second period, and may generate a delivery plan by a calculation using machine learning that matches the delivery timing to the users of the third group with the delivery timing to the users of the first or second group. In this case, by adjusting the delivery amount for users who are estimated to have a biomass fuel that exceeds the expiration date, it is possible to deliver the biomass fuel to the users of the first or second group without exceeding the expiration date. As a result, the disposal cost and delivery cost of the biomass fuel can be reduced. Effect of the Invention

[0008] According to a fuel delivery system according to one aspect of the present invention, it is possible to efficiently deliver fuel to a plurality of users having different consumption conditions, including the number of operating industrial vehicles and operating times. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating a fuel delivery system according to an embodiment. [Diagram 2] FIG. 2 is a block diagram showing an example of a hardware configuration of a delivery plan server. [Diagram 3] FIG. 2 is a block diagram showing an example of a configuration of an industrial vehicle. [Figure 4] FIG. 2 is a block diagram showing an example of a functional configuration of a delivery plan server. [Diagram 5] 13 is a flowchart illustrating an example of a process of a delivery plan server. [Figure 6] 6 is a flowchart showing a detailed example of the delivery plan generating process of FIG. 5. [Figure 7] FIG. 13 is a diagram showing an example of classification of users based on the evaluation result of consumption speed. [Figure 8]1A is a timing chart showing an example of a delivery plan for a first group of users, FIG. 1B is a timing chart showing an example of a delivery plan for a second group of users, and FIG. 1C is a timing chart showing an example of a consumption rate for a third group of users. [Figure 9] 1A is a timing chart showing another example of a delivery plan for a first group of users; FIG. 1B is a timing chart showing another example of a delivery plan for a second group of users; and FIG. 1C is a timing chart showing an example of generating a delivery plan for a third group of users as a second group of users. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] FIG. 1 is a diagram for explaining a fuel delivery system according to one embodiment. The fuel delivery system 1 is a system that generates a delivery plan for delivering fuel to a plurality of users 3 that use industrial vehicles 2 that run on fuel. One example of the industrial vehicle 2 is a forklift equipped with a diesel engine. The fuel here is a biomass fuel that is expected to have a CO2 reduction effect from the perspective of carbon neutrality, such as a biodiesel fuel such as fatty acid methyl ester (FAME). The industrial vehicle 2 is configured so that biodiesel fuel can be used for fuel system parts of the diesel engine, etc.

[0012] The fuel is produced by a fuel manufacturer 4. The fuel manufacturer 4 produces biodiesel fuel from waste vegetable oil and the like. The fuel manufacturer 4 may be a pre-selected company. Biodiesel fuel oxidizes more easily than fuels such as diesel oil. For this reason, a predetermined expiration date (e.g., three months) is set for it. To protect the diesel engines of the industrial vehicles 2, biodiesel fuel that has passed its expiration date is, in principle, not used but disposed of.

[0013] Fuel is delivered from the fuel manufacturer 4 to each of the multiple users 3 by a delivery vehicle 5. The delivery vehicle 5 delivers fuel to each of the multiple users 3 in accordance with a delivery plan described below. The delivery vehicle 5 delivers fuel to at least one of the multiple users 3 in one delivery. As the delivery vehicle 5, for example, a tanker truck 5a or a truck 5b carrying drums is used.

[0014] The multiple users 3 are provided with fuel tank equipment 6 for storing the delivered fuel until it is refueled into the industrial vehicles 2. For example, a tank 6a provided as a fixed facility or a delivered drum 6b is used as the tank equipment 6. The fuel delivered by the tank truck 5a is supplied to the tank 6a in a delivery amount according to a delivery plan described below. The drums 6b as the tank equipment 6 are delivered by the truck 5b in a number according to the delivery plan described below, and are kept at the user 3 until the next delivery timing.

[0015] [Fuel delivery system configuration] The configuration of the fuel delivery system 1 will be described below with reference to the drawings. As shown in Fig. 1, the fuel delivery system 1 includes a delivery plan server 10. The delivery plan server 10 is a server for generating a delivery plan.

[0016] The delivery plan server 10 is configured to be able to communicate with user terminals 3X of multiple users 3 and manufacturer terminals 4X of fuel manufacturers 4 via a network N. The delivery plan server 10 may be provided, for example, in a facility F that manages the industrial vehicles 2 of multiple users 3, or in a facility of the manufacturer of the industrial vehicles 2. The delivery plan server 10 may be composed of multiple computers provided in different locations. The delivery plan server 10 is connected to terminals 7 and a user management database 8.

[0017] The terminal 7 is a device for operating the delivery plan server 10. The terminal 7 may be, for example, a computer or a mobile terminal. The terminal 7 is not essential, and the delivery plan server 10 may be operated using an input device connected to the delivery plan server 10.

[0018] The user terminal 3X is a device for the user 3 to manage the industrial vehicle 2 and provide information to the delivery plan server 10. The user terminal 3X may be, for example, a computer or a mobile terminal.

[0019] The manufacturer terminal 4X is a device for managing fuel production in the fuel manufacturer 4 and acquiring information on delivery arrangements from the delivery plan server 10. The manufacturer terminal 4X may be, for example, a computer or a mobile terminal.

[0020] The user management database 8 is a database that stores information on the use of the industrial vehicle 2 for multiple users 3. The user management database 8 stores, for example, initially acquired information provided at the time of a fuel delivery contract with the user 3. The initially acquired information includes, for example, location information of the location where the industrial vehicle 2 of the user 3 is used, the capacity of the tank equipment 6 of the user 3, the number of industrial vehicles 2 deployed to the user 3, and the fuel tank capacity of each industrial vehicle 2. The user management database 8 also stores, for example, continuously acquired information transmitted from multiple users 3 via the network N. The continuously acquired information includes the number of operating industrial vehicles 2, the operating time of each industrial vehicle 2 (for example, on a daily basis), the fuel consumption amount of each industrial vehicle 2 (for example, on a daily basis), the remaining amount of fuel in the fuel tank for each industrial vehicle 2, the refueling timing for each industrial vehicle 2, the amount of refueling for each industrial vehicle 2, the fuel production date, the delivery date of the fuel, and the delivery amount of the fuel. The continuously acquired information may be associated with the identification information of the industrial vehicle 2. The identification information may be any information that can identify each individual industrial vehicle 2. The identification information may be an ID number or a vehicle number.

[0021] The hardware configuration of the delivery plan server 10 will be described. Fig. 2 is a block diagram showing an example of the hardware configuration of the delivery plan server. As shown in Fig. 2, the delivery plan server 10 is configured as a general computer equipped with a processor 10a, a storage unit 10b, a communication unit 10c, and a user interface 10d.

[0022] The processor 10a operates various operating systems to control the delivery plan server 10. The processor 10a is an arithmetic unit such as a CPU (Central Processing Unit) including a control device, an arithmetic device, a register, etc. The processor 10a controls the storage unit 10b, the communication unit 10c, and the user interface 10d. The storage unit 10b is a recording medium including at least one of a ROM (Read Only Memory), a RAM (Random Access Memory), a HDD (Hard Disk Drive), and an SSD (Solid State Drive).

[0023] The communication unit 10c is a communication device for performing wireless communication via the network N. A network device, a network controller, a network card, etc. can be used as the communication unit 10c. The delivery plan server 10 communicates with the user terminals 3X of the multiple users 3 and the manufacturer terminals 4X of the fuel manufacturers 4 using the communication unit 10c. The user interface 10d is an input / output unit of the delivery plan server 10 for the administrator of the delivery plan server 10, etc. The user interface 10d includes output devices such as a display and a speaker, and an input device such as a touch panel.

[0024] [Industrial vehicle configuration] 3 is a block diagram showing an example of the configuration of an industrial vehicle. The industrial vehicle 2 includes a controller 20, an engine load sensor 21, an engine speed sensor 22, and an airflow sensor 23 provided on a diesel engine, and an injector 24. Each sensor and the injector 24 are connected to the controller 20 that manages the overall control of the diesel engine.

[0025] The controller 20 is an electronic control unit having a CPU, a ROM, a RAM, etc. In the controller 20, for example, a program recorded in the ROM is loaded into the RAM, and the program loaded into the RAM is executed by the CPU, thereby realizing various functions. Note that the controller 20 may be composed of multiple electronic units.

[0026] The injectors 24 inject fuel into the combustion chambers of each cylinder of the diesel engine. The injectors 24 supply fuel into the combustion chambers by injecting the amount of fuel to be injected (hereinafter simply referred to as the injected fuel amount) and at the injection timing set by the controller 20. A common rail (not shown) is connected to each injector 24. The common rail stores high-pressure fuel to be supplied to each injector 24. The injection operation of each injector 24 is controlled by the controller 20.

[0027] The engine load sensor 21 is a detector that detects the load of the diesel engine, such as an accelerator opening sensor and a loading lever position sensor. The engine speed sensor 22 is a detector that detects the engine speed of the diesel engine. The air flow sensor 23 is a detector that detects the amount of intake air of the diesel engine. The engine load sensor 21, the engine speed sensor 22, and the air flow sensor 23 transmit detection signals of the detected engine state quantities to the controller 20.

[0028] The engine state quantity acquisition unit 20a acquires engine state quantities based on the detection results of various sensors. The engine state quantity acquisition unit 20a acquires a load command value for the diesel engine based on the detection results of the engine load sensor 21 and the engine speed sensor 22. The engine state quantity acquisition unit 20a calculates a load command value, which is a command value for the torque to be output by the diesel engine, based on, for example, the detection results of the accelerator opening sensor and the loading lever position sensor and the engine speed. The engine state quantity acquisition unit 20a acquires the intake air amount of the diesel engine based on.

[0029] The fuel injection control unit 20b calculates an injection fuel amount for making the diesel engine generate a combustion torque corresponding to the load command value based on, for example, the detection result of the air flow sensor 23 and the load command value of the diesel engine. The fuel injection control unit 20b controls the injector 24 to inject fuel in the calculated injection fuel amount.

[0030] The transmission information acquisition unit 20c acquires transmission information to be transmitted as continuously acquired information to the delivery plan server 10 via the user terminal 3X. The transmission information acquisition unit 20c acquires information corresponding to the industrial vehicle 2 from the continuously acquired information described above as transmission information. The transmission information acquisition unit 20c acquires, for example, the operating time of the industrial vehicle 2, the fuel consumption of the industrial vehicle 2, the remaining fuel tank amount of the industrial vehicle 2, the refueling timing of the industrial vehicle 2, and the amount of fuel refueled of the industrial vehicle 2 as transmission information. The refueling timing of the industrial vehicle 2 and the amount of fuel refueled of the industrial vehicle 2 may be manually input by the user 3 to the user terminal 3X.

[0031] The communication unit 20d is a communication device that controls wireless communication with the outside of the industrial vehicle 2. The communication unit 20d transmits and receives various information through communication with the user terminal 3X.

[0032] The information transmitting unit 20e transmits the transmission information to the user terminal 3X through the communication unit 20d. The information transmitting unit 20e transmits the transmission information to the user terminal 3X, for example, at the end of a day's operation of the industrial vehicle 2 and when refueling is performed. The information transmitting unit 20e may transmit the transmission information to the user terminal 3X at regular time intervals while the industrial vehicle 2 is in operation.

[0033] [Functional configuration and processing of the delivery plan server] Next, an example of the functional configuration and processing of the delivery plan server 10 will be described. Fig. 4 is a diagram showing an example of the functional configuration of the delivery plan server. Fig. 5 is a flowchart showing an example of processing of the delivery plan server.

[0034] As shown in FIG. 4, the delivery plan server 10 includes a consumption status acquisition unit 11, a delivery plan generation unit 12, and a delivery arrangement unit 13.

[0035] The consumption status acquisition unit 11 acquires the consumption status of each of the multiple users 3 (step S01). The consumption status is a parameter related to the fuel consumption rate of each of the multiple users 3. The consumption status includes, for example, the capacity of the fuel tank equipment 6, the remaining amount of fuel in the tank equipment 6, the number of operating industrial vehicles 2, the operating time of the industrial vehicles 2, and the fuel consumption amount of the industrial vehicles 2.

[0036] The consumption status acquisition unit 11 may acquire the consumption status based on, for example, the initial acquisition information and the continuously acquired information previously stored in the user management database 8. The consumption status acquisition unit 11 may directly acquire the consumption status based on the continuously acquired information acquired via the network N. The consumption status acquisition unit 11 continuously acquires (for example, on a daily basis) the continuously acquired information from each of the multiple users 3 for at least the operating time of each industrial vehicle 2, the fuel consumption amount of each industrial vehicle 2, and the remaining fuel tank amount of each industrial vehicle 2.

[0037] The consumption status acquisition unit 11 may calculate an average consumption status based on the acquired consumption status (step S02). The average consumption status is an average value of the consumption status in a predetermined period. The average consumption status includes, for example, the number of operating industrial vehicles 2, the average operating time of each industrial vehicle 2, the average fuel consumption of each industrial vehicle 2, the average number of refuelings of each industrial vehicle 2, the refueling frequency of each industrial vehicle 2, and the average fuel consumption of each user 3. The average fuel consumption of each user 3 means the average value of the total fuel consumption of a plurality of industrial vehicles 2 for the user 3. The predetermined period may be, for example, the past one week, the past one month, or the past three months for the average operating time and the average fuel consumption. The predetermined period may be, for example, the past one week for the average fuel consumption of each user 3.

[0038] The delivery plan generating unit 12 generates a delivery plan by calculation using machine learning based on the consumption status and the capacity of the tank facility 6 so as to optimize the delivery sequence, required time, and transportation cost for the multiple users 3 by calculation using machine learning (step S03). The delivery plan includes at least the next delivery timing (schedule), the delivery sequence for the multiple users 3, and the delivery amount for each user 3. When generating the delivery plan, the delivery style (whether direct replenishment by tank truck 5a or replacement of drum cans 6b by truck 5b) may be predetermined as part of the initially acquired information.

[0039] The delivery plan generating unit 12 generates a delivery plan by dividing the future period into a plurality of periods determined based on, for example, the fuel production date and the expiration date of the fuel. The future period is a period for which consumption prediction is performed based on the consumption status with the fuel production date as a reference. For example, if the expiration date of the fuel is three months, the future period may be divided into three periods: one month with the fuel production date as a reference, two months with the fuel production date as a reference, and three months with the fuel production date as a reference. The future period may be divided into more detailed periods, such as two weeks with the fuel production date as a reference, one month and two weeks with the fuel production date as a reference, and two months and two weeks with the fuel production date as a reference.

[0040] The delivery plan generating unit 12 generates a provisional delivery plan for each of the multiple users 3, dividing it into three future periods: one month from the fuel production date, two months from the fuel production date, and three months from the fuel production date. The fuel production date here means the production date of the fuel to be prepared in the next delivery arrangement, and may be a past production date for fuel that has already been produced, or a future production date for fuel that has not been produced at the time of generating the delivery plan but will be produced by the time the next delivery arrangement is made.

[0041] The delivery plan generating unit 12 generates a delivery plan by calculation using machine learning. For example, a genetic algorithm (GA) can be used as the machine learning. As an example, the machine learning here is configured to calculate the required time and transportation cost for each of a plurality of future periods based on the travel time, delivery work time, and cost unit price when assuming a delivery order for all or a portion of a plurality of users 3, and to increase the adaptability as the transportation cost decreases. The cost unit price may be preset as a logistics unit price per unit volume of fuel based on, for example, the transportation cost when delivering by tanker truck 5a, the transportation cost when delivering by truck 5b, and the size (maximum loading capacity) of tanker truck 5a and truck 5b.

[0042] For example, if the calculation result shows that the required time exceeds a predetermined maximum transport time, a penalty may be applied to lower the fitness. If the result of delivery of a certain delivery amount exceeds the capacity of the tank equipment 6 of the user 3, a penalty may be applied to lower the fitness. If the tank equipment 6 of the user 3 becomes depleted in the period between a certain delivery timing and the next delivery timing, a penalty may be applied to lower the fitness. The depleted state may be, for example, a state in which the remaining amount of the tank equipment 6 falls below a predetermined number of days of the predicted consumption amount (described later). If the expiration date of the fuel remaining in the tank equipment 6 of the user 3 expires, a penalty may be applied to lower the fitness. For example, when the number of days until the expiration date of the fuel remaining in the tank equipment 6 falls below the value obtained by adding a predetermined number of days to the number of days until the next delivery timing and the remaining amount of the tank equipment 6 exceeds the predetermined number of days of the predicted consumption, the fuel may be treated as having expired. Furthermore, when the post-delivery fuel amount, which is the sum of the remaining amount of fuel remaining in the tank facility 6 of the user 3 and the delivery amount at the next delivery timing, is taken as 100%, if the ratio of the remaining amount of fuel before delivery exceeds, for example, about 10%, a penalty may be given to reduce the fitness. In machine learning, other penalties may be appropriately set so that the fitness increases as the transportation cost decreases.

[0043] In addition, when the delivery plan generating unit 12 generates a provisional delivery plan, the delivery plan generating unit 12 may regenerate the delivery plan based on the continuously acquired information newly acquired after the generation of the delivery plan until the deadline for the delivery arrangement is reached. When the deadline for the delivery arrangement is reached, the delivery plan generating unit 12 may determine the latest delivery plan as the plan for the next delivery.

[0044] The delivery arrangement unit 13 arranges for the delivery of the fuel according to the determined delivery plan (step S04). For example, the delivery arrangement unit 13 places an order for the fuel and requests delivery to the manufacturer terminal 4X of the fuel manufacturer 4 via the network N according to the determined delivery plan. As a result, the fuel is delivered to each of the multiple users 3.

[0045] In this embodiment, the delivery plan generating unit 12 generates a delivery plan by machine learning calculation using delivery timing that is adjusted in advance so as not to increase the number of deliveries excessively. Figure 6 is a flowchart showing a detailed example of the delivery plan generating process of Figure 5. As shown in Figure 6, the delivery plan generating unit 12 evaluates the consumption speed of each of the multiple users 3 based on the consumption situation and the capacity of the tank facility 6 (step S11).

[0046] Specifically, the delivery plan generating unit 12 calculates, for example, the consumption rate for a predetermined future period (for example, three months based on the fuel production date in accordance with the expiration date of the fuel). The delivery plan generating unit 12 calculates, for example, a predicted consumption amount for one month (for example, corresponding to a one-month future period based on the fuel production date) based on an average consumption situation for a predetermined past period (for example, the past month). The predicted consumption amount for one month may be calculated, for example, by multiplying the average fuel consumption amount for each industrial vehicle 2 in the past month by the number of industrial vehicles 2 in operation. Since a three-month future period is assumed here, the consumption rate can be set to three times the predicted consumption amount for one month.

[0047] The delivery plan generating unit 12 uses the calculated consumption rate to evaluate the consumption rate of each of the multiple users 3. The evaluation of the consumption rate means an evaluation of how fast the consumption rate is relative to the capacity of the tank equipment 6. The evaluation of the consumption rate corresponds to an evaluation of whether the tank equipment 6 will reach a depleted state and whether the remaining fuel in the tank equipment 6 will pass its expiration date.

[0048] The delivery plan generating unit 12 evaluates the consumption rate, for example, according to the magnitude relationship between a numerical value (hereinafter referred to as a consumption rate evaluation value) obtained by dividing the calculated consumption rate (three times the predicted consumption amount for one month) by the capacity of the tank facility 6 and a predetermined threshold value. The predetermined threshold value is a consumption rate evaluation value threshold value for evaluating the consumption rate. For example, a first threshold value, a second threshold value smaller than the first threshold value, and a third threshold value smaller than the second threshold value may be used as the predetermined threshold value. The first threshold value may be, for example, a value of 3 or more (for example, about 10). The second threshold value may be, for example, a value of 1.5 or more and less than 3 (for example, about 2). The third threshold value may be, for example, a value less than 1.5 (for example, about 0.9).

[0049] The larger the predetermined threshold value, the faster the consumption rate relative to the capacity of the tank equipment 6, and the less likely it is that the remaining fuel in the tank equipment 6 will exceed its expiration date, but the more likely it is that the tank equipment 6 will run out of fuel.The smaller the predetermined threshold value, the slower the consumption rate relative to the capacity of the tank equipment 6, and the less likely it is that the tank equipment 6 will run out of fuel, but the more likely it is that the remaining fuel in the tank equipment 6 will exceed its expiration date.

[0050] The delivery plan generating unit 12 classifies the multiple users 3 into a first group and a second group based on the evaluation result of the consumption speed (step S12). FIG. 7 is a diagram showing an example of user classification based on the evaluation result of the consumption speed. As shown in FIG. 7, the delivery plan generating unit 12 classifies the users 3 whose consumption speed evaluation value is equal to or greater than a first threshold into the first group. The first group is a classification of users 3 who regularly deliver fuel at a predetermined first cycle.

[0051] FIG. 8(a) shows an example of a delivery plan for a first group of users 3. As shown in FIG. 8(a), fuel is delivered to the first group of users 3 at regular delivery timings of schedules T1, T2, T3, T4, T5, and T6. The intervals between the schedules T1, T2, T3, T4, T5, and T6 correspond to a predetermined first cycle. For the first group of users 3, fuel may be replenished from 10% to 100% of the remaining amount in the tank equipment 6 at each delivery timing. In the example of FIG. 8(a), the schedules T1, T2, T3, T4, T5, and T6 are at equal intervals, but depending on the result of generating a delivery plan based on continuously acquired information, they do not all have to be at equal intervals, and may be periodic.

[0052] The delivery plan generating unit 12 classifies the users 3 whose consumption speed evaluation value is less than the first threshold into a second group. The second group is a classification of users 3 who irregularly deliver fuel at a second cycle that is longer than the first cycle.

[0053] Fig. 8(b) shows an example of a delivery plan for the second group of users 3. As shown in Fig. 8(b), fuel is delivered to the second group of users 3 irregularly at a second cycle that is longer than the first cycle in which fuel is delivered to the first group of users 3. The delivery plan for the second group of users 3 will be described in detail later.

[0054] The delivery plan generating unit 12 classifies users 3 whose consumption speed evaluation value is less than the second threshold and equal to or greater than the third threshold into a third group. The third group is a classification of users 3 who irregularly receive fuel delivery in a second cycle that is longer than the first cycle, and who are estimated to have a second cycle that exceeds the expiration date if the next delivery volume is the first delivery volume. In other words, the delivery plan generating unit 12 further classifies, among the users 3 in the second group, users 3 who are estimated to have a second cycle that exceeds the expiration date if the next delivery volume is the first delivery volume into the third group.

[0055] FIG. 8(c) shows an example of the consumption rate of the third group of users 3. As shown in FIG. 8(c), when the third group of users 3 is replenished with fuel from 10% to 100% on the fuel production date, it is estimated that fuel will not be delivered for three months from the fuel production date, and fuel will remain in the tank equipment 6 at the time of three months from the fuel production date. Therefore, when the third group of users 3 is replenished with fuel at a delivery amount (first delivery amount) of 90% of the capacity of the tank equipment 6 as in the example of FIG. 8(c), it is estimated that the next delivery timing calculated from three months from the fuel production date will exceed three months (expiration date). In other words, the third group of users 3 are users 3 of the second group 3 whose second cycle is estimated to exceed the expiration date when the next delivery amount is the first delivery amount.

[0056] The delivery plan generating unit 12 determines whether or not a second group of users 3 exists (step S13). If it is determined that a second group of users 3 exists, the delivery plan generating unit 12 proceeds to step S14. If it is determined that a second group of users 3 does not exist, the delivery plan generating unit 12 proceeds to step S15.

[0057] In step S14, the delivery plan generating unit 12 matches the delivery timing for the second group of users 3 to the regular delivery timing for the first group of users 3. That is, the delivery plan generating unit 12 generates a delivery plan by calculation using machine learning that matches the delivery timing for the second group of users 3 to the regular delivery timing for the first group of users 3.

[0058] The second group of users 3 may irregularly receive fuel at delivery timings that do not coincide with any of the schedules T1 to T6. The delivery plan generating unit 12 matches the delivery timings to the second group of users 3 with the schedules T2 and T5 that are part of the regular delivery timings to the first group of users 3. For example, on the schedule T2, the remaining amount of the tank equipment 6 of the second group of users 3 has not yet reached 10%. However, if fuel is not delivered to the second group of users 3 on the schedule T2, the timing at which the remaining amount of the tank equipment 6 reaches 10% may be timing T0 that does not coincide with any of the schedules T1 to T6 (see the dashed line in FIG. 8(b)). Therefore, the delivery plan generating unit 12 may set the delivery timing to the second group of users 3 at a timing that coincides with the regular delivery timings to the first group of users 3 when, for example, the remaining amount of the tank equipment 6 of the second group of users 3 is equal to or less than a predetermined remaining amount (for example, equal to or less than half the capacity of the tank equipment 6). This allows the timing of delivery to the second group of users 3 to coincide with the regular delivery timing to the first group of users 3. As a result, in the example of Fig. 8(b), fuel is delivered to the second group of users 3 at the delivery timings of dates T2 and T5. For the second group of users 3, fuel is replenished from 15% to 100% at the delivery timing on date T2, and fuel is replenished from 10% to 100% at the delivery timing on date T5.

[0059] The delivery plan generating unit 12 determines whether or not a third group of users 3 exists (step S15). If it is determined that a third group of users 3 exists, the delivery plan generating unit 12 proceeds to step S16. The delivery plan generating unit 12 sets the next delivery amount for the third group of users 3 to a second delivery amount that is smaller than the first delivery amount and such that the second cycle does not exceed the expiration date (step S16).

[0060] Fig. 9(c) is a timing chart showing an example of generating a delivery plan with the third group of users 3 as the second group of users 3. The origin of Fig. 9 corresponds to three months from the fuel production date in Fig. 8. The origin of Fig. 9 is a fuel production date different from that in Fig. 8. Note that fuel is delivered to the first group of users 3 at regular delivery times on the dates T7, T8, T9, T10, T11, and T12 (see Fig. 9(a)).

[0061] As shown in FIG. 9(c), the third group of users 3 is supplied with fuel at a delivery amount (second delivery amount, here, delivery amount of 50% of the capacity of the tank equipment 6) until the remaining amount of the tank equipment 6 is 60% on the fuel production date of FIG. 9. In this case, the third group of users 3 is expected to consume fuel until the remaining amount of the tank equipment 6 is 30% one month after the fuel production date of FIG. 9 based on the average consumption situation. Therefore, one month after the fuel production date of FIG. 9, the third group of users 3 is supplied with fuel at a delivery amount (second delivery amount, delivery amount of 30% of the capacity of the tank equipment 6) until the remaining amount of the tank equipment 6 is 60%, similar to the fuel production date of FIG. 9. As a result, the third group of users 3 is expected to consume fuel until the remaining amount of the tank equipment 6 is 30% two months after the fuel production date of FIG. 9, similar to one month after the fuel production date of FIG. 9. After this, the same fuel consumption and delivery are repeated. As a result, the timing of delivery to the third group of users 3 will coincide with the regular delivery timing to the first group of users 3 on schedule T8, and will coincide with the delivery timing to the first and second groups of users 3 on schedule T10.

[0062] In this way, the delivery plan generating unit 12 adjusts the delivery amount for the third group of users 3 to the second delivery amount, thereby matching the delivery timing for the third group of users 3 to the delivery timing for the first or second group of users (step S17). That is, the delivery plan generating unit 12 sets the next delivery amount for the third group of users 3 to the second delivery amount that is smaller than the first delivery amount and the second cycle does not exceed the expiration date, and generates a delivery plan by a calculation using machine learning that matches the delivery timing for the third group of users 3 to the delivery timing for the first or second group of users 3. As a result, it becomes possible to change and handle the third group of users 3 who are part of the second group of users 3 to the second group of users 3 who do not actually belong to the third group.

[0063] On the other hand, if it is determined in step S15 that there are no users 3 in the third group, the delivery plan generating unit 12 ends the processing in FIG. 6 and returns to step S04 in FIG.

[0064] According to the fuel delivery system 1 described above, the consumption rates of the multiple users 3 are evaluated based on the consumption status and the capacity of the tank equipment 6. Based on the evaluation result of the consumption rate, the multiple users 3 are classified into a first group to which fuel is delivered periodically at a predetermined first cycle, and a second group to which fuel is delivered irregularly at a second cycle longer than the first cycle. A delivery plan is generated by a calculation using machine learning that matches the delivery timing to the second group of users 3 with the regular delivery timing to the first group of users 3. As a result, the number of deliveries is reduced and the increase in delivery costs can be suppressed compared to a delivery plan in which, for example, delivery to the first group of users 3 and delivery to the second group of users 3 are performed separately. Therefore, according to the fuel delivery system 1, it is possible to efficiently deliver fuel to multiple users 3 with different consumption statuses including the number of industrial vehicles 2 in operation and the operating time.

[0065] In the fuel delivery system 1, the fuel is a biomass fuel that is discarded after the expiration date, and the delivery plan generation unit 12 further classifies the users 3 of the second group, who are estimated to have a second period that exceeds the expiration date when the next delivery amount is set to the first delivery amount, into a third group based on the evaluation result of the consumption rate, and sets the next delivery amount for the users 3 of the third group to a second delivery amount that is smaller than the first delivery amount and does not exceed the expiration date in the second period, and generates a delivery plan by a calculation using machine learning that matches the delivery timing to the users 3 of the third group with the delivery timing to the users 3 of the first or second group. As a result, by adjusting the delivery amount for the users 3 who are estimated to have a biomass fuel that exceeds the expiration date, it is possible to deliver the biomass fuel to the users 3 of the first or second group without exceeding the expiration date. As a result, it is possible to reduce the disposal cost and delivery cost of the biomass fuel.

[0066] Although the embodiment of the present invention has been described above, the present invention is not limited to the above-mentioned embodiment. The present invention can be embodied in various forms including the above-mentioned embodiment and various modifications and improvements based on the knowledge of those skilled in the art.

[0067] In the above embodiment, the fuel is biomass fuel that is discarded after the expiration date, but this is not limited to this example. For example, even if the fuel is diesel, the delivery plan generation unit may classify multiple users into a first group and a second group based on the evaluation result of the consumption rate, and generate a delivery plan by calculation using machine learning that matches the delivery timing to the users of the second group with the regular delivery timing to the users of the first group.

[0068] In the above embodiment, the delivery plan generating unit further classifies the users in the second group, who are estimated to have their second period exceed the expiration date when the next delivery amount is the first delivery amount, into the third group, but is not limited to this example. For example, the classification into the third group may be omitted.

[0069] In the above embodiment, the multiple users 3 included users 3 who use tanker trucks 5a as delivery vehicles 5, users 3 who use trucks 5b as delivery vehicles 5, and users 3 who use both tanker trucks 5a and trucks 5b as delivery vehicles 5, but it is not necessary to include all three of these. The multiple users 3 may be made up only of users 3 who use tanker trucks 5a as delivery vehicles 5. The multiple users 3 may be made up only of users 3 who use trucks 5b as delivery vehicles 5. In these cases, the delivery plan generation unit can generate a delivery plan to coordinate delivery timing between the multiple users 3 who share a common delivery vehicle 5. [Explanation of symbols]

[0070] 1...fuel delivery system, 2...industrial vehicle, 3...user, 6...tank equipment, 11...consumption status acquisition unit, 12...delivery plan generation unit.

Claims

1. A fuel delivery system that generates a delivery plan for delivering the fuel to a plurality of users who use industrial vehicles operated by fuel, For each of the plurality of users, a consumption status acquisition unit that acquires a consumption status including the capacity of the fuel tank facility, the remaining amount of the tank facility, the number of operating industrial vehicles, the operating time of the industrial vehicle, and the fuel consumption of the industrial vehicle; A delivery plan generation unit that generates the delivery plan so as to optimize the delivery order, required time, and transportation cost of the plurality of users by calculation using machine learning based on the consumption status and the capacity of the tank facility, The delivery plan generation unit, Evaluates the consumption rates of the plurality of users based on the consumption status and the capacity of the tank facility, Classifies the plurality of users into a first group that regularly delivers the fuel at a predetermined first cycle and a second group that irregularly delivers the fuel at a second cycle longer than the first cycle based on the evaluation result of the consumption rate, Using the calculation using machine learning that matches the delivery timing to the users in the second group with the regular delivery timing to the users in the first group, with the regular delivery timing to the users in the first group, the consumption status of the users in the second group, and the capacity of the tank facility of the users in the second group as inputs, a fuel delivery system that generates the delivery plan in which the delivery timing to the users in the second group is made to coincide with any of the regular delivery timings to the users in the first group.

2. The fuel is biomass fuel that is discarded when the expiration date has passed, The delivery plan generation unit, Based on the evaluation result of the consumption rate, among the users in the second group, further classifies the users for whom it is estimated that the second cycle exceeds the expiration date when the next delivery amount is set as the first delivery amount into a third group, The fuel delivery system according to claim 1, wherein the next delivery amount for the users in the third group is set as a second delivery amount that is smaller than the first delivery amount and the second period does not exceed the expiration date, and the delivery plan is generated by the calculation using the machine learning in which the delivery timing to the users in the third group is adjusted to match the delivery timing to the users in the first group or the second group.

Images