Home presence rate management device and home presence rate management method

Abstract

To provide a home presence rate management device and a home presence rate management method which can precisely predict the home presence rate of a resident.SOLUTION: A home presence rate management device includes a home presence rate prediction information storage unit and a home presence rate prediction information processing unit. The home presence rate prediction information storage unit has a home presence rate prediction data table for storing the prediction value of the home presence rate set for each house and each time zone. The home presence rate prediction information processing unit acquires information of the home presence rate and the home absence rate for each house and each time zone as information on delivery operations that a distributor has done in a predetermined period of time, and adds more predicted values of corresponding houses in corresponding time zones in the home presence rate prediction data table when the amount of information showing presence in the home is larger, and subtracts more predicted values when the amount of information showing absence in the home is larger, on the basis of the acquired information.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a home occupancy rate management device and a home occupancy rate management method.

Background Art

[0002] Conventionally, when a logistics company delivers a package to an individual's residence, based on past experience, it may predict the occupancy rate of residents on weekdays and time periods, and conduct delivery operations aiming at the time when the occupancy rate is high.

[0003] In addition, in collective housing such as condominiums, there may be a case where information linking the vehicle in / out information of a mechanical parking lot and the parking lot contract information is managed. By providing this information to a logistics company, the logistics company can accurately predict the in-home / out-of-home status of residents in each residence, and can preferentially deliver packages to residences where the possibility of being at home is high (the vehicle is stored in the parking lot). Also, the logistics company can suppress the occurrence of redelivery items by not delivering packages during that time period to residences where the possibility of residents being at home is low (the vehicle is not stored in the parking lot).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, even if a resident is using a vehicle and it is not stored in the parking lot, there may be a case where another resident (family member) is at home in the corresponding residence, or even if a vehicle is stored in the parking lot, all the residents of the residence may be out without using the vehicle. There has been a problem that it is difficult to accurately predict the occupancy rate only based on the vehicle in / out information of the parking lot.

[0006] The present invention has been made in view of the above circumstances, and an object thereof is to provide an occupancy rate management device and an occupancy rate management method capable of accurately predicting the occupancy rate of residents at home.

Means for Solving the Problems

[0007] The occupancy rate management device of the present invention for achieving the above object includes an occupancy rate prediction information storage unit that holds an occupancy rate prediction data table storing predicted values of the occupancy rate set in advance for each residence and each time zone, and information on delivery operations performed by a logistics company within a predetermined period, as information on presence or absence at home for each residence and each time zone. Based on the acquired information, for the predicted value of the occupancy rate of the corresponding residence in the corresponding time zone stored in the occupancy rate prediction data table, a larger value is added as the information on presence is more, and a larger value is subtracted as the information on absence is more. An occupancy rate prediction information processing unit.

[0008] Further, the occupancy rate management method of the present invention is an occupancy rate management device including an occupancy rate prediction information storage unit that holds an occupancy rate prediction data table storing predicted values of the occupancy rate set in advance for each residence and each time zone. As information on delivery operations performed by a logistics company within a predetermined period, information on presence or absence at home for each residence and each time zone is acquired, and based on the acquired information, for the predicted value of the occupancy rate of the corresponding residence in the corresponding time zone stored in the occupancy rate prediction data table, a larger value is added as the information on presence is more, and a larger value is subtracted as the information on absence is more.

Effects of the Invention

[0009] According to the occupancy rate management device and the occupancy rate management method of the present invention, the occupancy rate of residents owning vehicles can be accurately predicted.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Embodiments for Carrying Out the Invention

[0011] 〈Configuration of a Home Occupancy Rate Prediction System According to an Embodiment〉 The configuration of a home occupancy rate prediction system according to an embodiment of the present invention will be described with reference to FIG. 1. The home occupancy rate prediction system 1 according to this embodiment is a system for predicting the home occupancy rate for each of the residences C1, C2, C3,... in an apartment house A such as a condominium, and includes a parking lot management device 10, a delivery business operator management device 20 connected to the parking lot management device 10, and a delivery person terminal 30 connected to the delivery business operator management device 20.

[0012] The parking lot management device 10 is a device that manages information regarding the parking lot P of the apartment complex A. The parking lot P has a plurality of parking spaces Q1, Q2, Q3...

[0013] The delivery service provider management device 20 is a device that manages information regarding the delivery business for a delivery person X who delivers delivered items to residences within the apartment complex A, and the delivery person terminal 30 is a terminal carried by a delivery person of the delivery service provider X. In FIG. 1, only one delivery person terminal 30 is shown for simplicity of explanation, but actually, a plurality of delivery person terminals carried by a plurality of delivery persons are connected to the delivery service provider management device 20.

[0014] The parking lot management device 10 includes a user registration processing unit 11, a registration information database (DB) 12, an in / out information acquisition unit 13, a first home occupancy rate prediction information processing unit 14, a first home occupancy rate prediction information storage unit 15, and a first communication unit 16.

[0015] The user registration processing unit 11 performs a process of inputting, as information of a user who has contracted to use the parking lot P, a customer ID and name which are identification information of the user, address information (room number within the apartment complex A), family composition, identification information of the owned vehicle and vehicle type information, and identification information of the parking space that is the contract target. The registration information DB 12 registers the user information input by the user registration processing unit 11. The in / out information acquisition unit 13 acquires the in / out information of vehicles for each of the parking spaces Q1, Q2, Q3... in the parking lot P.

[0016] The first home occupancy rate prediction information processing unit 14 generates a home occupancy rate prediction data table that stores initial values of predicted home occupancy rates preset for each residence C1, C2, C3..., for each day type classified based on the usage pattern of the parking lot, and for each predetermined time period. As the day types, for example, "weekdays" which are days for commuting or going to school, "holidays" which correspond to weekends or single-day holidays, long consecutive holidays such as New Year's and events, and "special days" where the usage pattern of the parking lot is predicted to be in a special state such as the days before them are set. In this embodiment, the case where "weekdays" and "holidays" are set as the day types will be described.

[0017] Then, when the first home occupancy rate prediction information processing unit 14 detects that a vehicle has entered any of the parking spaces based on the information acquired by the in-out information acquisition unit 13, it identifies the user who has the right to use the parking space based on the information stored in the registration information DB 12. The first home occupancy rate prediction information processing unit 14 temporarily increases the predicted value of the home occupancy rate for the time period after detecting the entry of the vehicle of the corresponding day type in the home occupancy rate prediction data table regarding the residence of the corresponding user, such that the shorter the elapsed time since the detection, the higher the value, and transmits the changed content to the delivery service provider management device 20.

[0018] Also, when the first home occupancy rate prediction information processing unit 14 detects that a vehicle has left any of the parking spaces based on the information acquired by the in-out information acquisition unit 13, it changes the predicted value of the home occupancy rate for the corresponding day type and time period regarding the residence of the corresponding user in the home occupancy rate prediction data table to decrease, and transmits the changed content to the delivery service provider management device 20.

[0019] Also, the first home occupancy rate prediction information processing unit 14 updates the generated home occupancy rate prediction data table based on the information regarding the delivery service acquired from the delivery service provider management device 20.

[0020] The first home occupancy rate prediction information storage unit 15 stores the home occupancy rate prediction data table generated by the first home occupancy rate prediction information processing unit 14. The first communication unit 16 performs information communication with the delivery service provider management device 20.

[0021] The delivery service provider management device 20 includes a second communication unit 21, a delivery information acquisition unit 22, a second home occupancy rate prediction information processing unit 23, and a second home occupancy rate prediction information storage unit 24.

[0022] The second communication unit 21 performs information communication with the parking lot management device 10. The delivery information acquisition unit 22 acquires, from the delivery person terminal 30, information indicating "success" (when the resident is at home in the corresponding residence) or "failure" (when the resident is not at home in the corresponding residence), which is the result information of the delivery acquired when the delivery service of delivery service provider X is executed by the delivery person.

[0023] The second occupancy rate prediction information processing unit 23 acquires the information received by the parking lot management device 10 via the second communication unit 21, stores it in the second occupancy rate prediction information storage unit 24, and changes or updates it as necessary. Further, the second occupancy rate prediction information processing unit 23 transmits the information acquired by the delivery information acquisition unit 22 to the parking lot management device 10. Further, the second occupancy rate prediction information processing unit 23 provides the information stored in the second occupancy rate prediction information storage unit 24 to the delivery person terminal 30 in response to a request from the delivery person terminal 30.

[0024] <Operation of the Occupancy Rate Prediction System Using the Occupancy Rate Prediction Information Generation Device According to an Embodiment> Next, as the operation of the occupancy rate prediction system 1 according to the present embodiment, (1) the process executed at the time of the parking lot use contract procedure, (2) the process of temporarily changing the occupancy rate prediction data table according to the use status of the parking space, and (3) the process of updating the occupancy data table according to the delivery status of the home delivery company will be described with reference to the drawings.

[0025] (1) Process Executed at the Time of the Parking Lot Use Contract Procedure First, the process executed by the occupancy rate prediction system 1 when a new use contract procedure is performed by the residents of the dwelling in the apartment complex A to use the parking lot P will be described with reference to the flowchart of FIG. 2.

[0026] When a new use contract procedure for the parking lot P is performed by the user E who is a resident of the dwelling C1 in the apartment complex A (''YES'' in S1), the user registration processing unit 11 of the parking lot management device 10 uses the personal information of the user E as the customer ID, name, address information (room number), family composition (age, driver's license possession information, occupation, etc.), identification information of the owned vehicle and vehicle type information, and the information of the parking space Q1 which is the identification information of the parking space to be contracted, and registers it in the registration information DB12 (S2).

[0027] Next, the first home occupancy rate prediction information processing unit 14 acquires information regarding the user E registered in the registration information DB 12, and based on the acquired information, stores initial values of predicted values of the home occupancy rate for each day type and each time zone regarding the residence C1 of the user E, and generates a home occupancy rate prediction data table as home occupancy rate prediction information (S3). The home occupancy rate prediction data table generated here is information that serves as a basis for the predicted value of the home occupancy rate of the residence C1 of the user E.

[0028] An explanation will be given of the generation process of the home occupancy rate prediction data table executed by the first home occupancy rate prediction information processing unit 14. The first home occupancy rate prediction information processing unit 14 holds in advance a home occupancy rate initial data table in which initial parameters of the home occupancy rate are set for each predetermined time zone for weekdays / holidays respectively. An example of the home occupancy rate initial data table is shown in FIG. 3.

[0029] In the home occupancy rate initial data table of FIG. 3, for weekdays / holidays respectively, initial parameters of the home occupancy rate assumed based on the general living patterns of humans are set every 30 minutes between the times of 7:00 and 21:00. These initial parameters assume that on weekdays, many people go to work around 7:00 without using a vehicle and return home around 18:00, and on holidays, vehicles are often used when going out, and if a vehicle is parked in the corresponding parking space, it is highly likely that the user is at home. Thus, the predicted values of the home occupancy rate for each time zone are set in the range of 80% to 20%.

[0030] Based on the home occupancy rate initial data table held by the first home occupancy rate prediction information processing unit 14, the personal information of the user E registered in the registration information DB 12, the information on the usage frequency of the vehicle of the user E for each weekday / holiday, the regional characteristic information of the apartment complex A, etc., the home occupancy rate prediction data table of the user E is generated.

[0031] The information on the usage frequency of the vehicle of the user E may be obtained from the user E by questionnaire at the time of the usage contract procedure for the parking space Q1, or the usage status of the parking space Q1 of the user E for a certain period may be obtained and the information aggregated for each weekday / holiday may be used.

[0032] In addition, the regional characteristic information of the apartment house A is, for example, information indicating whether the distance or walking time from the apartment house A to the nearest station or bus stop exceeds a preset threshold value k, the timetable of public transportation, and the like.

[0033] First, the first occupancy rate prediction information processing unit 14 determines which of the following attributes (i) and (ii) the dwelling C1 of the user E corresponds to according to the living habits of the user E.

[0034] For attribute (i), since the resident can easily go out without using a vehicle, uses public transportation for commuting, and is likely to be absent when using a vehicle, when a vehicle is parked in the corresponding parking space, the relevance between the utilization status of the parking space and the occupancy rate is low, and when no vehicle is parked in the corresponding parking space, the dwelling where the relevance between the utilization status of the parking space and the occupancy rate is high corresponds.

[0035] In addition, for attribute (ii), since the resident often uses a vehicle when going out and also uses a vehicle for commuting, the dwelling where the relevance between the utilization status of the parking space and the occupancy rate is high corresponds regardless of whether a vehicle is parked in the parking space.

[0036] The first occupancy rate prediction information processing unit 14 determines which of these attributes (i) and (ii) the dwelling of the user E corresponds to based on the personal information of the user E, the information on the usage frequency of the vehicle for each weekday / holiday of the user E, the regional characteristic information of the apartment house A, etc., and classifies the dwelling C1 of the user E into the corresponding attribute.

[0037] For example, based on the response content of the questionnaire of user E or the usage status over a certain period, the corresponding attributes are judged and classified. Or, based on the regional characteristic information of the residence of user E, when the distance or walking time from the apartment building A to the nearest station or bus stop exceeds the threshold value k, the residence of user E is classified as attribute (ii). Also, based on the timetable of public transportation, for example, a train, when the number of trains per hour during the day (for example, 7:00 to 18:00) is equal to or more than the reference value S, the possibility of using the train is high and the frequency of using a vehicle is low. Therefore, the residence of user E is classified as attribute (i).

[0038] When classifying the residence of user E in this way, the first occupancy rate prediction information processing unit 14 corrects the predicted value of the occupancy rate in the occupancy rate initial data table based on the information used for the classification determination, and generates an occupancy rate prediction data table for the residence of the user E.

[0039] For example, based on the response content of the questionnaire obtained from user E at the time of the usage contract procedure for the parking space Q1 and the usage status over a certain period, for weekdays / holidays respectively, the predicted value of the occupancy rate for the time periods with a high possibility of being at home is increased, and the predicted value of the occupancy rate for the time periods with a low possibility of being at home is decreased.

[0040] At this time, regarding the residence classified as attribute (i), when a vehicle is stored in the corresponding parking space, the relevance between the usage status of the parking space and the occupancy rate is low. Therefore, the correction coefficient of the occupancy rate is lowered.

[0041] An example of the occupancy rate prediction data table for the residence of user E generated by correcting the values in the occupancy rate initial data table is shown in FIG. 4. In this occupancy rate prediction data table, compared with the values in the occupancy rate initial data table of FIG. 3, the occupancy rate is corrected to be lower mainly during the day.

[0042] The first occupancy rate prediction information processing unit 14 stores the generated occupancy rate prediction data table for user E in the first occupancy rate prediction information storage unit 15 and transmits it to the delivery company management device 20 via the first communication unit 16 (S4).

[0043] The home delivery service provider management device 20 receives, via the second communication unit 21, the occupancy rate prediction data table of the residence of user E transmitted from the parking lot management device 10, and the second occupancy rate prediction information processing unit 23 stores it in the second occupancy rate prediction information storage unit 24 (S5).

[0044] (2) Process of temporarily changing the occupancy rate prediction data table according to the usage status of the parking space Next, a process of temporarily changing the occupancy rate prediction data table of the residence of user E according to the usage status of the parking space Q1 used by user E will be described with reference to the flowchart of FIG. 5.

[0045] The first occupancy rate prediction information processing unit 14 acquires the entry / exit information via the entry / exit information acquisition unit 13 each time a vehicle enters or exits in the parking spaces Q1, Q2, Q3,... of the parking lot P. Then, when the first occupancy rate prediction information processing unit 14 acquires the entry / exit information of the parking space Q1 (”YES” in S11), it temporarily changes the occupancy rate prediction data table regarding the residence C1 of user E stored in the first occupancy rate prediction information storage unit 15 (S12).

[0046] A process of temporarily changing the occupancy rate prediction data table of the residence of user E will be described. When the vehicle is not in the corresponding parking space (when the vehicle is empty), the possibility of the corresponding residence being occupied decreases. Therefore, the first occupancy rate prediction information processing unit 14 lowers the predicted value of the corresponding time period in the occupancy rate prediction data table by a predetermined ratio.

[0047] Here, the first occupancy rate prediction information processing unit 14 holds, in advance, an initial value of “0.5” of a correction coefficient (correction coefficient at the time of vehicle exit) for lowering the predicted value when the vehicle is not in the corresponding parking space. In addition, it holds a vehicle type correction table storing a correction coefficient (correction coefficient by vehicle type) for further adjusting this correction coefficient at the time of vehicle exit according to the vehicle type of the vehicle.

[0048] An example of a vehicle type correction table is shown in FIG. 6. In the vehicle type correction table of FIG. 6, for coupes, small cars, and light automobiles, the number of passengers is 1 to 2, and they are likely to be owned by single-person or small-sized households. When there is no vehicle in the parking space for such a vehicle, it is predicted that the occupancy rate of the corresponding residence is low, and the correction coefficient for each vehicle type of these vehicles is "0.8". Also, for sedans, since they are likely to be owned by family households of three or more people, even if there is no vehicle in the parking space for such a vehicle, it is predicted that there may be family members at home in the corresponding residence, and the correction coefficient for each vehicle type of sedans is "1". Also, for wagons, one-box vehicles, and sport utility vehicles (SUVs), since they are likely to be owned by large-family households, on weekdays, even if there is no such vehicle in the parking space, it is predicted that there may be family members at home in the corresponding residence, and the correction coefficient for each vehicle type of the corresponding vehicle on weekdays is "1". Also, for wagons, one-box vehicles, and SUVs, on holidays, when there is no corresponding vehicle in the parking space, it is predicted that the occupancy rate of the corresponding residence is low, and the correction coefficient for each vehicle type of the corresponding vehicle on holidays is "0.8".

[0049] Also, when the vehicle enters the parking space it corresponds to, the first occupancy rate prediction information processing unit 14 holds an in-stock correction table that stores a correction coefficient (in-stock correction coefficient) for increasing the predicted value for a predetermined time from the time of entry.

[0050] An example of the in-stock correction table is shown in FIG. 7. In the in-stock correction table of FIG. 7, the added value for the predicted value up to 3 hours after every 30 minutes from the time of entry is stored. The added value at the time of entry is "0%", the added value for the time period from the time of entry to 0.5 hours after entry is "90%", the added value for the time period from 0.5 hours after entry to 1.0 hours after entry is "75%", the added value for the time period from 1.0 hours after entry to 1.5 hours after entry is "60%", the added value for the time period from 1.5 hours after entry to 2.0 hours after entry is "45%", the added value for the time period from 2.0 hours after entry to 2.5 hours after entry is "30%", and the added value for the time period from 2.5 hours after entry to 3.0 hours after entry is "15%".

[0051] Each time the first home occupancy rate prediction information processing unit 14 acquires the inbound and outbound information regarding any vehicle space, it uses these outbound correction coefficients, vehicle type correction table, and inbound correction table to change the predicted value of the home occupancy rate for the corresponding time period in the home occupancy rate prediction data table stored in the first home occupancy rate prediction information storage unit 15.

[0052] For example, when the first home occupancy rate prediction information processing unit 14 detects that a vehicle in parking space Q1 has left at 9:30 on a weekday, it identifies user E who has the right to use the parking space Q1 based on the information stored in the registration information DB12, and changes the predicted value of the home occupancy rate for the time period corresponding to the time when the departure is detected on a weekday for the residence C1 of user E based on the outbound correction coefficient and the vehicle type correction table. Thereafter, until a vehicle enters the parking space Q1, the predicted value of the home occupancy rate for the corresponding time period is maintained at the value changed when the departure was detected.

[0053] Also, when the first home occupancy rate prediction information processing unit 14 detects that a vehicle has entered the parking space Q1 at 13:00 on a weekday, it changes the predicted value of the home occupancy rate for the time period from 13:00 to 16:00, which corresponds to 3 hours after the entry is detected on a weekday for the residence C1 of user E, based on the value in the inbound correction table. Each time the first home occupancy rate prediction information processing unit 14 changes the predicted value due to the inbound and outbound of the vehicle, it transmits the changed home occupancy rate prediction data table to the courier management device 20 via the first communication unit 16 (S12).

[0054] The courier management device 20 receives the changed home occupancy rate prediction data table of user E's residence transmitted from the parking lot management device 10 via the second communication unit 21, and updates the home occupancy rate prediction data table of user E stored in the second home occupancy rate prediction information storage unit 24 based on the information received by the second home occupancy rate prediction information processing unit 23 (S13). The processing of steps S11 to S13 is repeated until the date changes ("NO" in S14).

[0055] The first occupancy rate prediction information processing unit 14 shows an example of an occupancy rate prediction data table changed within one day based on the in-out status of vehicles in the parking lot P for weekdays and holidays respectively in FIG. 8.

[0056] The occupancy rate prediction data table in FIG. 8 shows the predicted values of the occupancy rate on the day when the vehicle of user E left the warehouse at 9:30 on a weekday and entered the warehouse at 13:00, and on the day when it left the warehouse at 14:00 on a holiday and entered the warehouse at 17:00.

[0057] Here, the vehicle of user E is a sedan, the correction coefficient at the time of leaving the warehouse uses the initial value "0.5" as it is, and in the time periods from 9:30 to 13:00 on weekdays and from 14:00 to 17:00 on holidays when the parking space Q1 is in an empty state, the predicted values in the occupancy rate prediction data table in FIG. 4 are changed to the values obtained by multiplying each predicted value by 0.5.

[0058] Also, the predicted values in the time periods from 3 hours after 13:00 on weekdays when the vehicle entered the warehouse until 3 hours after 17:00 on holidays when the vehicle entered the warehouse are changed to the values obtained by adding the corresponding added values for each time period to the predicted values in the occupancy rate prediction data table in FIG. 4 (however, the maximum value is 100%).

[0059] Specifically, in the time period from 13:00 to 13:30 on weekdays, the added value "90%" for the time period from when the vehicle entered the warehouse until 0.5 hours after entering the warehouse is added to each predicted value "15%" in the occupancy rate prediction data table in FIG. 4, and the predicted value of the occupancy rate is changed to the maximum value of "100%". Similarly, in the time period from 13:30 to 14:00 on weekdays, the predicted value is "90%", in the time period from 14:00 to 14:30, the predicted value is "75%", in the time period from 14:30 to 15:00, the predicted value is "60%", in the time period from 15:00 to 15:30, the predicted value is "45%", and in the time period from 15:30 to 16:00, the predicted value is changed to "30%".

[0060] Also, during the time periods of 17:00 to 17:30, 17:30 to 18:00, and 18:00 to 18:30 on holidays, and during the time period of 18:30 to 19:00, the predicted value of the occupancy rate is changed to the highest value of "100%". During the time period of 19:00 to 19:30, the predicted value is "95%", and during the time period of 19:30 to 20:00, the predicted value is changed to "83%".

[0061] When a delivery person of courier company X performs an operation to request the predicted value of the occupancy rate of a desired user from the courier company management device 20 on the delivery person terminal 30 during the delivery operation, the second occupancy rate prediction information processing unit 23 acquires the predicted value of the occupancy rate of the corresponding user from the second occupancy rate prediction information storage unit 24 and provides it to the delivery person terminal 30. By obtaining the latest predicted value of the occupancy rate appropriately changed as described above, the delivery person can efficiently perform the delivery operation.

[0062] When the second occupancy rate prediction information processing unit 23 provides the information on the predicted value of the occupancy rate to the delivery person terminal 30, as shown in FIG. 9, a diagram showing the change over time in a day as a line graph for the predicted values for each time period on weekdays and holidays in the occupancy rate prediction data table may be added. The solid line in the graph indicates the change in the predicted value on weekdays, and the dotted line indicates the change in the predicted value on holidays.

[0063] Also, the occupancy rate prediction data table in FIG. 10 shows the predicted values of the occupancy rate on the day when user E's vehicle left the warehouse at 9:30 on a weekday and entered the warehouse at 17:00, and on the day when it left the warehouse at 10:00 on a holiday and entered the warehouse at 15:00. FIG. 11 is a diagram showing the predicted values for each time period on weekdays and holidays in the occupancy rate prediction data table changed as in FIG. 10 as a line graph. The solid line in the graph indicates the change in the predicted value on weekdays, and the dotted line indicates the change in the predicted value on holidays.

[0064] Comparing Fig. 9 and Fig. 11, even if the predicted occupancy rate is changed by the same method based on the same occupancy rate prediction data table of Fig. 4, if the times of stockout and stockin are different, the change in the predicted value (the slope of the broken line) is different. For example, when a vehicle is stocked in during a time period with a high occupancy rate in the occupancy rate prediction data table of Fig. 4, the possibility of going out later is low, and even if the added value decreases over time after stocking in, the predicted value does not decrease significantly.

[0065] After that, when the first occupancy rate prediction information processing unit 14 detects that the date has changed ("YES" in S14), it returns the values in the occupancy rate prediction data table changed during the day in the occupancy rate prediction data table stored in the first occupancy rate prediction information storage unit 15 to the state before being changed based on the inbound / outbound information, that is, the values shown in Fig. 4, and transmits them to the courier management device 20 via the first communication unit 16 (S15).

[0066] The courier management device 20 receives the changed occupancy rate prediction data table of the residence of user E transmitted from the parking lot management device 10 via the second communication unit 21, and the second occupancy rate prediction information processing unit 23 updates the occupancy rate prediction data table of user E stored in the second occupancy rate prediction information storage unit 24 (S16).

[0067] (3) Update process of the occupancy data table executed according to the delivery status of courier X Next, the process of updating the occupancy rate prediction data table of the residence of user E according to the delivery status of courier X will be described with reference to the flowchart of Fig. 12.

[0068] In this embodiment, each time a delivery person of a courier conducts a delivery operation, information indicating "success" (when there is an occupant in the corresponding residence) or "failure" (when the corresponding residence is unoccupied) as delivery result information regarding the residence of the delivery destination is transmitted from the delivery person terminal 30 to the courier management device 20. The courier management device 20 acquires the delivery result information transmitted from the delivery person terminal 30 by the delivery information acquisition unit 22.

[0069] When the delivery information acquisition unit 22 acquires the delivery result information ( "YES" in S21), the second at-home rate prediction information processing unit 23 transmits the acquired information to the parking lot management device 10 via the second communication unit 21 (S22).

[0070] In the parking lot management device 10, the first at-home rate prediction information processing unit 14 acquires the delivery result information transmitted from the home delivery service provider management device 20. Based on the acquired delivery result information, the first at-home rate prediction information processing unit 14 updates, for example, at the timing when the date changes, the corresponding daily type (weekday / holiday) and time zone at-home rate prediction data table regarding the residence of user E stored in the first at-home rate prediction information storage unit 15. This at-home rate prediction data table is in a state where no temporary prediction value change is made by the process of (2) above.

[0071] The process of updating the at-home rate prediction data table for the residence of user E will be described. The first at-home rate prediction information processing unit 14 preliminarily holds "+1%" as the addition value of the predicted at-home rate value when the delivery is "successful" and "-1%" as the subtraction value when the delivery is "failed". Also, when correcting the predicted value by adding the addition value to the predicted value or subtracting the subtraction value from the predicted value, it holds a weighting value table storing the weighting values for these addition and subtraction values based on the in-vehicle / vacant state of the corresponding parking space.

[0072] An example of the weighting value table is shown in FIG. 13. In the weighting value table of FIG. 13, when delivery is successful in the vehicle-present state where there is a vehicle in the corresponding parking space, the addition value is multiplied by 1.5, that is, “+1.5%”, and when delivery fails in the vehicle-present state, the subtraction value remains multiplied by 1.0, that is, “-1.0%”. When delivery is successful in the vehicle-absent state where there is no vehicle in the corresponding parking space, the addition value remains multiplied by 1.0, that is, “+1.0%”, and when delivery fails in the vehicle-absent state, the subtraction value is multiplied by 1.5, that is, “-1.5%”. Values indicating this are stored in this weighting value table. In this weighting value table, when delivery is successful in the vehicle-present state and when delivery fails in the vehicle-absent state, it is predicted that the presence or absence of a vehicle has a great influence on the occupancy rate, and the correction weight is increased.

[0073] Each time the first occupancy rate prediction information processing unit 14 acquires the delivery result information, it weights the addition value or subtraction value based on the delivery result information with the value stored in the weighting value table and stores it as a prediction correction value. Then, for each preset time period, the stored prediction correction values are totaled.

[0074] Here, the preset time periods for totaling the prediction correction values are time periods divided for each time when the occupancy rate is considered to be approximated due to living habits. For example, there are five time periods: the AM1 time period from 7:00 to 10:00 am, the AM2 time period from 10:00 to 12:00 am, the PM1 time period from 12:00 to 15:00 pm, the PM2 time period from 15:00 to 18:00 pm, and the PM3 time period from 18:00 to 21:00 pm on weekdays and holidays respectively.

[0075] During the AM1 time period on weekdays, it is a time period when the occupancy rate at home is predicted to decrease due to going to work. The AM2 time period, PM1 time period, and PM2 time period are time periods when people are at work and the occupancy rate at home is predicted to be low. The PM3 time period is a time period when the occupancy rate at home is predicted to increase due to returning home. Also, during the AM1 time period on holidays, it is a time period when many people who get up later than on weekdays are at home and the occupancy rate is predicted to be high. The PM2 time period is a time period when the occupancy rate is predicted to decrease because the possibility of going out increases. The PM1 time period is a time period when there are both people resting at home and people going out, and the occupancy rate is predicted to be about 50%. The PM2 time period is a time period when people who went out are gradually starting to return home and the occupancy rate is predicted to increase. The PM3 time period is a time period when even more people return home and the occupancy rate is predicted to increase.

[0076] For each of these time periods, examples of the total value of the prediction correction values based on the delivery result information generated regarding the residence of user E within a predetermined learning period (for example, one month) are shown in FIGS. 14 and 15.

[0077] FIG. 14(a) is an example of the total value of the prediction correction values based on the delivery result information generated during the corresponding period for the AM1 time period on weekdays and holidays respectively regarding the residence of user E. During the corresponding time period on weekdays, when there is a vehicle parked in parking space Q1, the delivery by courier X was successful once, and based on this delivery result information, the total value of the prediction correction value "+1.5%" is stored. Also, during the corresponding time period on weekdays, when there is an empty vehicle in parking space Q1, the delivery by courier X failed once, and based on this delivery result information, the total value of the prediction correction value "-1.5%" is stored.

[0078] Also, during the corresponding time period on holidays, when there is a vehicle in parking space Q1, the delivery by courier X has been successful 2 times (+1.5% × 2) and failed 1 time (-1%). Based on this delivery result information, the total value of the predicted correction values obtained by summing these, which is "2%", is stored. Also, during the corresponding time period on holidays, when the vehicle in parking space Q1 is empty, the delivery by courier X has failed 2 times (-1.5% × 2) and been successful 1 time (1%). Based on this delivery result information, the total value of the predicted correction values, which is "-2%", is stored.

[0079] Similarly, Fig. 14(b) shows an example of the value obtained by summing the predicted correction values based on the delivery result information that occurred during the corresponding period for each of the AM2 time periods on weekdays and holidays for the residence of user E. Also, Fig. 15(a) shows an example of the value obtained by summing the predicted correction values based on the delivery result information that occurred during the corresponding period for each of the PM1 time periods on weekdays and holidays for the residence of user E. Also, Fig. 15(b) shows an example of the value obtained by summing the predicted correction values based on the delivery result information that occurred during the corresponding period for each of the PM2 time periods on weekdays and holidays for the residence of user E. Also, Fig. 15(c) shows an example of the value obtained by summing the predicted correction values based on the delivery result information that occurred during the corresponding period for each of the PM3 time periods on weekdays and holidays for the residence of user E.

[0080] The delivery result correction value table storing the total value of the predicted correction values for each time period of in-vehicle / empty-vehicle for each of these weekdays and holidays generated by the first home occupancy rate prediction information processing unit 14, and the value obtained by further summing up the total values of the predicted correction values for each time period of in-vehicle / empty-vehicle for each time period, is shown in FIG. 16. In the delivery result correction value table of FIG. 16, the total value of the predicted correction values for the AM1 time period on weekdays is "0.0%", the total value of the predicted correction values for the AM2 time period on weekdays is "0.0%", the total value of the predicted correction values for the PM1 time period on weekdays is "0.0%", the total value of the predicted correction values for the PM2 time period on weekdays is "0.0%", the total value of the predicted correction values for the PM3 time period on weekdays is "-7.5%", the total value of the predicted correction values for the AM1 time period on holidays is "0.0%", the total value of the predicted correction values for the AM2 time period on holidays is "2.5%", the total value of the predicted correction values for the PM1 time period on holidays is "-8.5%", the total value of the predicted correction values for the PM2 time period on holidays is "0.0%", and the total value of the predicted correction values for the PM3 time period on holidays is "-2.5%" are stored.

[0081] Based on the generated delivery result correction value table, an example of the home occupancy rate prediction data table updated by the first home occupancy rate prediction information processing unit 14 for each of weekdays and holidays is shown in FIG. 17.

[0082] In the home occupancy rate prediction data table of FIG. 17, the predicted values for each time period in the home occupancy rate prediction data table of FIG. 4 are updated by being corrected with the corresponding values in the delivery result correction value table of FIG. 16. Here, the correction is performed by rounding up the values after the decimal point in the delivery result correction value table.

[0083] When the same result (delivery success / failure) is repeated a predetermined number of times under the same conditions (weekday / holiday and time period), the first home occupancy rate prediction information processing unit 14 may increase the degree of correction of the predicted value for the corresponding conditions, assuming that the result has reproducibility.

[0084] For example, when the same result is repeated five or more times under the same conditions using the delivery result information generated during the PM3 time period on weekdays and holidays respectively for the residence of user E over a predetermined period, the case where the absolute value of the prediction correction value after the fifth time is increased by 20% will be described. In the said delivery result information, since success has continued five times when the vehicle is in the parked state on weekdays, as shown in FIG. 18, the absolute value of the prediction correction value for the fifth time is increased by 20% from "1.5%" to "1.8%". Also, afterwards, since failure has continued five times, the absolute value of the prediction correction value for the fifth time is increased by 20% from "-1%" to "-1.2%". Further, since failure has continued eight times when the vehicle is in the empty state on weekdays, the absolute value of the prediction correction value from the fifth to the eighth time is changed from "-1.5%" to "-1.8%".

[0085] Similarly, since success has continued nine times when the vehicle is in the parked state on holidays, the absolute value of the prediction correction value from the fifth to the ninth time is increased by 20% from "1.5%" to "1.8%". Also, since failure has continued ten times when the vehicle is in the empty state on holidays, the absolute value of the prediction correction value from the fifth to the ninth time is changed from "-1.5%" to "-1.8%", and for the tenth time it is further decreased to "-2.1%".

[0086] In this way, the higher the degree that the same result continues in the same time period, the more corrections are added to the prediction correction value, so that the high reproducibility can be accurately reflected in the predicted value of the occupancy rate and updated.

[0087] After that, when the first occupancy rate prediction information processing unit 14 acquires the vehicle entry and exit information to the parking lot P, using the updated occupancy rate prediction data table in the first occupancy rate prediction information storage unit 15, as described in (2), the change process of the predicted value of the occupancy rate for the corresponding time period is executed.

[0088] Also, the first occupancy rate prediction information processing unit 14 may perform a temporary update process based on the delivery result information on the occupancy rate prediction data table for which a temporary change has been made to the predicted value during the day.

[0089] At that time, in the process of (2) above, since the predicted value of the occupancy rate is changed according to the occupied / vacant state of the parking space, the first occupancy rate prediction information processing unit 14 obtains the added value of the corresponding state (occupied / vacant) from the weighting value table in FIG. 13 for the occupancy rate prediction data table to which a temporary prediction value change is added, and performs an update process based on the delivery result information.

[0090] For example, by performing an update process based on the delivery result information on the occupancy rate prediction data table temporarily changed as shown in FIG. 10, an occupancy rate prediction data table as shown in FIG. 19 is generated. By performing an update process based on the delivery result information on the occupancy rate prediction data table temporarily changed according to the occupied / vacant state of the parking space in this way, the occupancy rate on that day can be predicted with higher accuracy.

[0091] When the first occupancy rate prediction information processing unit 14 updates the occupancy rate prediction data table stored in the first occupancy rate prediction information storage unit 15 based on the delivery result information, it transmits the updated information to the parcel delivery company management device 20 (S23).

[0092] When the parcel delivery company management device 20 receives the updated occupancy rate prediction data table of the residence of user E transmitted from the parking lot management device 10, the second occupancy rate prediction information processing unit 23 updates the occupancy rate prediction data table of user E stored in the second occupancy rate prediction information storage unit 24 (S24). The processes in steps S21 to S24 are repeatedly executed for a predetermined learning period (for example, one month).

[0093] In the above-described embodiment, since the time zone when the parcel delivery company X makes a delivery may be a time zone with a high probability of occupancy specified by the resident of the corresponding residence, the first occupancy rate prediction information processing unit 14 may count the number of deliveries for each time zone on weekdays and holidays, and correct the predicted value in the occupancy rate prediction data table based on the count value.

[0094] For example, as shown in FIG. 20, the first home occupancy rate prediction information processing unit 14 counts the number of deliveries that occurred during the AM1 hour period, AM2 hour period, PM1 hour period, PM2 hour period, and PM3 hour period within a predetermined period related to the residence of user E for each weekday and holiday. In the example of FIG. 20, the number of weekday deliveries during the AM1 hour period is 2, the number of holiday deliveries is 6, the number of weekday deliveries during the AM2 hour period is 20, the number of holiday deliveries is 12, the number of weekday deliveries during the PM1 hour period is 8, the number of holiday deliveries is 16, the number of weekday deliveries during the PM2 hour period is 10, the number of holiday deliveries is 2, the number of weekday deliveries during the PM3 hour period is 5, and the number of holiday deliveries is 20, and they are counted as such.

[0095] Then, based on the counted values, as shown in FIG. 21, the first home occupancy rate prediction information processing unit 14 obtains the median value "8" of the number of deliveries for all time periods on weekdays and the median value "12" of the number of deliveries for all time periods on holidays, converts the difference between the counted value for each time period and the corresponding median value into a percentage value, and calculates it as a prediction correction value based on the number of deliveries. In the example of FIG. 21, the prediction correction value for weekdays during the AM1 hour period is "-6.0%", the prediction correction value for holidays is "-6.0%", the prediction correction value for weekdays during the AM2 hour period is "12.0%", the prediction correction value for holidays is "0.0%", the prediction correction value for weekdays during the PM1 hour period is "0.0%", the prediction correction value for holidays is "4.0%", the prediction correction value for weekdays during the PM2 hour period is "2.0%", the prediction correction value for holidays is "-10.0%", the prediction correction value for weekdays during the PM3 hour period is "-3.0%", and the prediction correction value for holidays is "8.0%", and they are calculated as such.

[0096] Then, the first at-home rate prediction information processing unit 14 corrects the corresponding predicted value in the at-home rate prediction data table based on the calculated prediction correction value. For example, as shown in FIG. 22, the first at-home rate prediction information processing unit 14 reduces each predicted value corresponding to the weekday AM1 time zone by 6.0% from the initial value, increases each predicted value corresponding to the weekday AM2 time zone by 12.0%, increases each predicted value corresponding to the weekday PM2 time zone by 2.0%, and reduces each predicted value corresponding to the weekday PM3 time zone by 3.0%. Also, it reduces each predicted value corresponding to the holiday AM1 time zone by 6.0%, increases each predicted value corresponding to the holiday PM1 time zone by 4.0%, reduces each predicted value corresponding to the holiday PM2 time zone by 10.0%, and reduces each predicted value corresponding to the holiday PM3 time zone by 8.0%.

[0097] FIG. 23 is a line graph showing the changes over time during a day for the predicted values (initial values and corrected values) for each time zone on weekdays and holidays respectively in the at-home rate prediction data table corrected as described above. The thick dotted line in the graph shows the change in the initial value of the predicted value on weekdays, the thin dotted line shows the change in the initial value of the predicted value on holidays, the thick solid line shows the change in the corrected value on weekdays, and the thin solid line shows the change in the corrected value on holidays. For each of weekdays and holidays, based on the delivery record to the residence of user E, the higher the number of deliveries, the higher the predicted value of the at-home rate for the corresponding time zone, and the lower the number of deliveries, the lower the predicted value of the at-home rate for the corresponding time zone.

[0098] In addition, the first at-home rate prediction information processing unit 14 acquires the operation information of public transportation, and when it acquires information that there are delays or cancellations in public transportation, it may predict that the at-home rate of the residences classified into attribute (i) will increase, and temporarily increase the predicted value of the at-home rate of the corresponding residences.

[0099] The above-mentioned processes (1) to (3) are also executed for the residences of the users of parking lot P other than user E, and appropriate corrections are added to the predicted values in the at-home rate initial data table to generate a housing rate data table for the residences of the said users.

[0100] For example, regarding the residence C2 of user F who uses the parking space Q2 classified into attribute (ii), the first home occupancy rate prediction information processing unit 14 corrects the predicted value of the home occupancy rate in the home occupancy rate initial data table based on the answer content of the questionnaire of user F and the usage status over a certain period, and generates a home occupancy rate prediction data table as shown in FIG. 24. At this time, regarding the residences classified into attribute (ii), regardless of whether a vehicle is stored in the parking space, since the relevance between the usage status of the parking space and the home occupancy rate is high, the correction coefficient of the home occupancy rate is increased for all time zones.

[0101] Then, each time a vehicle enters or exits the parking space Q2, the first home occupancy rate prediction information processing unit 14 temporarily changes the home occupancy rate prediction data table for the corresponding time zone based on the entry / exit information acquired from the entry / exit information acquisition unit 13. FIG. 25 shows an example of the home occupancy rate prediction data table changed within one day based on the entry / exit status of the vehicle to the parking space Q2 on weekdays and holidays respectively by the first home occupancy rate prediction information processing unit 14. FIG. 26 is a diagram showing, in a line graph, the changes over time during a day regarding the predicted values after the change for each time zone on weekdays and holidays in the home occupancy rate prediction data table changed as described above. The solid line in the graph shows the change in the predicted value on weekdays, and the dotted line shows the change in the predicted value on holidays.

[0102] In addition, the first home occupancy rate prediction information processing unit 14 can further appropriately update the home occupancy rate prediction data table based on the delivery result information by the delivery company X regarding the residence C2 of the user F.

[0103] According to the above embodiments, it is possible to accurately predict the home occupancy rate of a residence using the usage status information of the parking lot with which a predetermined user has a usage contract and the delivery result information of the parcels to the residence of the user. By using the information on the home occupancy rate predicted in this way, the delivery company can accurately grasp the residences with a high home occupancy rate for each time zone and can efficiently perform the delivery business.

[0104] Also, in the above-described embodiment, generally, the going-out rate decreases during rainfall, and the utilization rate of vehicles during going out increases. Therefore, when the first occupancy rate prediction information processing unit 14 acquires weather information and information indicating that it is rainfall, it temporarily increases the predicted value of the occupancy rate of the dwelling corresponding to the parking space where the vehicle is stored (in-vehicle state), and temporarily decreases the predicted value of the occupancy rate of the dwelling corresponding to the parking space where the vehicle is not stored (empty-vehicle state) for correction.

Explanation of Signs

[0105] 1 Occupancy Rate Prediction System 10 Parking Lot Management Device 11 User Registration Processing Unit 12 Registration Information Database (DB) 13 In / Out Information Acquisition Unit 14 First Occupancy Rate Prediction Information Processing Unit 15 First Occupancy Rate Prediction Information Storage Unit 16 First Communication Unit 20 Courier Company Management Device 21 Second Communication Unit 22 Delivery Information Acquisition Unit 23 Second Occupancy Rate Prediction Information Processing Unit 24 Second Occupancy Rate Prediction Information Storage Unit 30 Courier Terminal

Claims

1. a home-at-home ratio prediction information storage unit that holds a home-at-home ratio prediction data table that stores a predicted value of the home-at-home ratio that is preset for each residence and each time period; and an at-home rate prediction information processing unit that acquires information on whether a person is at home or not for each residence and time period as information on delivery operations performed by a logistics company within a specified period, and based on the acquired information, adds a larger value to the predicted value of the at-home rate for the relevant time period for the relevant residence stored in the at-home rate prediction data table the more the at-home information there is, and subtracts a larger value the more the absent information there is.

2. a home-at-home ratio prediction information storage unit that holds a home-at-home ratio prediction data table that stores a predicted value of the home-at-home ratio that is preset for each residence and each time period; and an at-home rate prediction information processing unit that acquires information on the number of deliveries for each residence and time period as information regarding delivery operations performed by a logistics company within a specified period, and based on the acquired information, adds a higher value to the predicted at-home rate for the corresponding residence and time period stored in the at-home rate prediction data table the more the number of deliveries.

3. a home-at-home ratio management device including a home-at-home ratio prediction information storage unit that holds a home-at-home ratio prediction data table that stores a home-at-home ratio prediction value that is preset for each residence and each time period; This is an at-home rate management method, which obtains information on whether a person is at home or not for each residence and time period as information on delivery operations performed by a logistics company within a specified period, and based on the obtained information, adds a larger value to the predicted at-home rate for the relevant residence and for the relevant time period stored in the at-home rate prediction data table the more at-home information there is, and subtracts a larger value the more absent information there is.

4. a home-at-home ratio management device including a home-at-home ratio prediction information storage unit that holds a home-at-home ratio prediction data table that stores a home-at-home ratio prediction value that is preset for each residence and each time period; This is an at-home rate management method, which obtains information on the number of deliveries for each residence and time period as information on delivery operations performed by a logistics company within a specified period, and based on the obtained information, adds a higher value to the predicted at-home rate for the corresponding residence and time period stored in the at-home rate prediction data table the more the number of deliveries.

Images