Method for predicting daily available capacity of parcel pick-up station and computerized locker bank

Abstract

To provide a method for predicting daily available capacity of a parcel pick-up station and computerized locker banks.SOLUTION: In a parcel locker bank management system including at least one computerized parcel locker bank 50, a prediction unit that communicates with a locker bank computer 51 included in the parcel locker bank, and a destination determination unit that determines destinations of multiple parcels, the locker bank computer 51 includes a processor and a memory, obtains data related to the parcel locker bank 50, and transmits the data to the prediction unit. Each of parcel lockers 1-L to 20-M capable of individually ensuring security includes a door and a lock mechanism.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention generally relates to a luggage locker bank, which is also known as a luggage collection point. Specifically, the present invention relates to predicting the available capacity of a luggage locker bank.

Background Art

[0002] One of the main peculiarities of delivery to locations outside the home, namely luggage locker banks, is the fact that it is completely uncertain when the luggage will be picked up from the luggage locker where it has been deposited. Therefore, it is impossible to know whether the luggage locker containing the luggage to be picked up is available or whether it can be used for other deliveries. This uncertainty is, firstly, a problem for the carrier agent when the carrier agent needs to plan the delivery of luggage to the luggage locker bank. This is because it is inefficient to carry the luggage to the luggage locker bank and then possibly not be able to deposit the luggage in that luggage locker bank, and moreover, it takes time to wait to obtain information that the pick-up has ended and the luggage locker is available, which is also inefficient. This uncertainty is, secondly, a problem when the customer places an order for the luggage. This is because it is difficult, and in some cases almost impossible, to inform the customer of a specific location and pick-up time.

[0003] Furthermore, due to the above uncertainty regarding when the pick-up will be, a time lag occurs between when the luggage is deposited in the luggage locker and when the luggage is finally picked up by the recipient. This time lag results in a stock situation. Each piece of luggage is deposited for a finite period of unknown length.

[0004] The solution of reservation is the safest but not the most optimal. Carriers cannot propose a refined solution other than the reservation solution.

[0005] Document US2018 / 190062 describes a computerized locker network comprising multiple locker banks, multiple portable computing devices, and at least one logistics server, wherein each locker bank comprises a locker bank computer and a locker compartment that is individually secure and has a door and locking mechanism. This comprehensive system is configured to identify a locker compartment in a locker bank that is available to hold a package after an attempt to deliver it to the customer's home address was unsuccessful.

[0006] Therefore, improved systems and methods are needed to maximize the efficiency of the parcel locker bank by minimizing the risk of non-delivery. [Overview of the project]

[0007] The present invention aims to provide a computerized implementation method and a computerized system that helps maximize the efficiency of a package locker bank by predicting the daily available capacity of the package locker bank and minimizing the risk of undelivered packages.

[0008] The first subject of the present invention is a computer implementation method for predicting the daily available capacity of a luggage locker bank, wherein the locker bank comprises a plurality of individually (selectively) secure luggage lockers, and the method proposes a computer implementation method that allows luggage to be deposited in one of the plurality of luggage lockers of the luggage locker bank for a maximum number of days equal to a first integer, hereafter denoted as N.

[0009] This method comprises the following multiple prediction steps for a number of luggage lockers equal to a second integer: -Receive the number of packages delivered or scheduled to be delivered to the package locker bank for each day between the selected future date and N-1 days prior to the selected future date, where the selected date corresponds to the date on which the user wishes to have at least one package delivered to the package locker bank. -Estimate a third number corresponding to the number of parcel lockers that should be holding parcels on the above-selected future day, where the above estimate is based on the above-mentioned number of parcels received that have been delivered or are scheduled to be delivered, and the pick-up rates of parcels from parcel lockers, where the pick-up rates are given for each day and depend on the number of days a parcel which is still in a parcel locker has been stored for. -Calculate a fourth number corresponding to the number of available luggage lockers in the above luggage locker bank on the above selected future day, where the above fourth number is obtained by subtracting the above third number from the above second integer.

[0010] The methods described above can be applied to the flow of packages from any origin. For example, they can be applied to a second delivery of a package when the destination of the package is changed to a package locker bank after an initial attempt at delivery to a home. For example, they can be applied to determining the destination of packages (direct parcels) when two parties conducting e-commerce use package lockers to exchange packages. They can also be applied to the handover of packages when a merchant places a package in a locker for a delivery person to pick up.

[0011] The retrieval rate may be based on a training group of data regarding retrieval rates previously observed in other luggage locker banks. The above retrieval rate derived from the training data group may be used to initiate the luggage locker bank's methods and systems.

[0012] When the luggage locker bank is in use, the retrieval rate may be updated daily using daily data, which includes data on luggage retrieved on that day and luggage lockers where luggage is still available, as described in the following paragraphs.

[0013] In the first aspect of the computer implementation method, each pickup rate is estimated daily by regression during closing hours of the luggage locker bank. The regression takes into account the real pickup rates for a number of previous days for which the pickup rate is known, and whether some of those days are holidays.

[0014] The primary advantage of using rates is their aggregate nature. When considering two lockers, we can calculate the retrieval rate for each machine, but we can also calculate the retrieval rate for both lockers. This aggregate nature avoids the need to calculate separate retrieval rates for every locker and every pair of items.

[0015] Individual models are applied to large-capacity storage banks, while aggregated models are sufficient for others, and no overflow problems arise.

[0016] Furthermore, in daily regressions, in order to make the take-up rate as close as possible to the actual current take-up rate, the most recent data obtained regarding take-up may be given greater weight during the regression than older data.

[0017] In a second aspect of the computer implementation method, the prediction steps may further include a verification step after the calculation of the fourth number described above. The verification step may include comparing the fourth number to a threshold and indicating that there are no available luggage lockers if the fourth number is smaller than the threshold described above.

[0018] By comparing the fourth number mentioned above with the confirmation threshold, a higher degree of certainty can be obtained regarding the availability of luggage lockers on the desired day. In reality, since the availability check is based on probabilities of retrieval, there is a risk that some luggage lockers may not be available as predicted and may ultimately be unavailable. This threshold helps to reduce the risk of encountering such a situation.

[0019] In other embodiments, the computer implementation method may, after the calculation step, include a second calculation of a fourth number for each day between the above future date and the last day on which the parcel can be deposited, and a verification step. The fourth number may correspond to the number of available parcel lockers for each day of this period. The fourth number may be based on a third number and the received number of parcels delivered or announced to be delivered by carriers to parcel lockers of said specific type of the parcel locker bank on the day for which the calculation is made. The verification step may include comparing each calculated fourth number to a threshold and indicating that there are no available parcel lockers if each calculated fourth number is less than the threshold.

[0020] In a third aspect of the computer implementation method, the above luggage locker bank comprises a first class of luggage lockers of a first size and a second class of at least one luggage locker of a second size that is larger than the first size. The above multiple prediction steps are repeated for each class of luggage lockers. In each class of luggage lockers, the second integer corresponds to the number of luggage lockers of that class.

[0021] The prediction process for each class of luggage lockers can be performed simultaneously.

[0022] In a fourth aspect of the computer implementation method, the above luggage locker bank has a first group of luggage lockers assigned only to the first user, and a second group of luggage lockers assigned to the first user and at least one other user. The above multiple prediction steps are repeated for each group of luggage lockers. For each group of luggage lockers, the second integer corresponds to the number of luggage lockers in that group.

[0023] The prediction process for each group of luggage lockers can be performed simultaneously.

[0024] In a fifth aspect of the computer implementation method, the above luggage locker bank has a first group of luggage lockers assigned only to the first user, and a second group of luggage lockers assigned to the first user and at least one other user. Each group comprises a first group of luggage lockers of a first size, and a second group of at least one luggage locker of a second size larger than the first size. The above multiple prediction steps are repeated for each group in each group of luggage lockers. With respect to each group of luggage lockers, the second integer corresponds to the number of luggage lockers in the said lot for the corresponding group.

[0025] The execution of the forecasting steps for each size of each lot of parcel lockers can be performed simultaneously.

[0026] In a sixth aspect of the computer-implemented method, when a parcel locker bank includes defective parcel lockers, the second integer is recalculated taking into account not to include the number of defective parcel lockers.

[0027] A second subject matter of the present invention is a computer-implemented method for directing the destination of one or more parcels to an appropriate parcel locker bank, the method comprising: - receiving a request to deposit at least one parcel at one of a plurality of parcel locker banks located in the vicinity of the destination address on a selected future date; - selecting the parcel locker bank closest to the destination address and determining whether there is an appropriate parcel locker available for the parcel on the selected future date by executing the computer-implemented method described above for predicting the daily available capacity of the parcel locker bank; - if there is no appropriate parcel locker available for the parcel at the closest parcel locker bank, repeating the selection and prediction steps the required number of times until an appropriate parcel locker available on the selected future date is found, where the parcel locker bank in each iteration is the parcel locker bank closest to the destination address next to the previously selected parcel locker bank; - transmitting an indication of the parcel locker bank having an appropriate parcel locker available for depositing the parcel on the selected future date and querying whether the available appropriate parcel locker should be reserved; We propose a computer implementation method that includes the following features.

[0028] A third subject of the present invention is a luggage locker bank management system comprising at least one computerized luggage locker bank, the luggage locker bank comprising a plurality of individually secureable luggage lockers, allowing luggage to be deposited in one of the plurality of luggage lockers of the luggage locker bank for a maximum number of days equal to a first integer hereafter denoted as N, the luggage locker bank further comprising at least one locker bank computer, the at least one locker bank computer comprising at least one processor and memory, configured to obtain data relating to the luggage locker bank and transmit the data to an external party, the number of luggage lockers being equal to a second integer, each of the plurality of individually secureable luggage lockers comprising at least one door and at least one locking mechanism, the luggage locker bank management system comprising a prediction unit communicating with the locker bank computer of the luggage locker bank, the prediction unit being: - A receiving module configured to receive the number of packages delivered or scheduled to be delivered to the package locker bank for each day between a selected future date and a number of days equal to N-1 prior to the selected future date, wherein the selected date corresponds to a date on which the user wishes to have at least one package delivered to the package locker bank. - A first calculation module configured to estimate a third number corresponding to the number of luggage lockers that will contain luggage on the above-selected future day, wherein the above estimate is based on the above-mentioned number of received luggage that has been delivered or is scheduled to be delivered, and the luggage retrieval rate from luggage lockers, wherein the above retrieval rate depends on the number of days that luggage still stored in the luggage lockers has been stored up to that point. - A second calculation module configured to calculate a fourth number corresponding to the number of available luggage lockers in the above luggage locker bank on the above selected future day, where the above fourth number is the above second integer minus the above third number. We propose a luggage locker bank management system equipped with the following features.

[0029] In the first aspect of the luggage locker bank management system, the first calculation module described above comprises a regression module. The regression module is configured to estimate each retrieval rate by regression every day, outside of the luggage locker bank's operating hours. The regression considers the actual retrieval rates for several past days that are known from information obtained from actual retrieval dates for luggage deposited in the past, and takes into account whether or not holidays are included in those several days.

[0030] In a second aspect of the luggage locker management system, the calculation module may include a verification module. The verification module may be configured to compare a fourth number with a threshold and indicate that there are no available luggage lockers if the fourth number is less than the threshold.

[0031] In a third aspect of the luggage locker bank management system, the above luggage locker bank comprises a first class of luggage lockers of a first size and a second class of at least one luggage locker of a second size that is larger than the first size. The above multiple prediction steps are repeated for each class of luggage lockers. In each class of luggage lockers, the second integer corresponds to the number of luggage lockers of that class.

[0032] In the fourth aspect of the luggage locker bank management system, the above luggage locker bank has a first group of luggage lockers assigned only to the first user, and a second group of luggage lockers assigned to the first user and at least one other user. The above multiple prediction steps are repeated for each group of luggage lockers. For each group of luggage lockers, the second integer corresponds to the number of luggage lockers in that group.

[0033] In the fifth aspect of the luggage locker bank management system, the luggage locker bank has a first group of luggage lockers assigned only to the first user, and a second group of luggage lockers assigned to the first user and at least one other user. Each group comprises a first group of luggage lockers of a first size, and a second group of at least one luggage locker of a second size that is larger than the first size. The above prediction steps are repeated for each group of luggage lockers in each group of luggage lockers. With respect to each group of luggage lockers, the second integer corresponds to the number of luggage lockers in that group for that group.

[0034] In the sixth aspect of the luggage locker bank management system, if a luggage locker bank has malfunctioning luggage lockers, the second integer is recalculated taking into consideration that the number of malfunctioning luggage lockers is not included.

[0035] A seventh aspect of the luggage locker bank management system further comprises a destination determination unit configured to determine the destination of one or more pieces of luggage to the appropriate luggage locker bank, the destination determination unit being: - A request module configured to receive a request to deposit at least one piece of luggage on a selected future day in one of several luggage locker banks located near the destination address. - A selection module configured to select the baggage locker bank closest to the destination address, and to work with a prediction unit to determine whether there is a suitable baggage locker available for the baggage on the selected future date, and if there is no suitable baggage locker available for the baggage in the closest baggage locker bank, to repeat the process for the required number of times until a suitable baggage locker is found on the selected future date, wherein the baggage locker bank in each iteration is the baggage locker bank closest to the destination address after the previously selected baggage locker bank. - A transmission module configured to send a message indicating the above luggage locker bank that has a suitable luggage locker available for depositing the above luggage on the above selected future date, and to inquire whether or not to reserve the above available suitable luggage locker. It is equipped with.

[0036] The present invention will be better understood by reading the following specification, which is illustrative and not restrictive, with reference to the accompanying drawings. [Brief explanation of the drawing]

[0037] [Figure 1] Figure 1 is a schematic diagram of an example of a luggage locker bank. [Figure 2] Figure 2 is a flowchart of a computer implementation method according to an embodiment of the present invention for predicting the daily available capacity of a luggage locker bank. [Figure 3] Figure 3 is a flowchart of a computer implementation method according to an embodiment for determining the destination of one or more packages to the appropriate luggage locker bank in a luggage locker bank network. [Figure 4] Figure 4 is a schematic diagram of a luggage locker bank management system, which includes at least one computerized luggage locker bank as shown in Figure 1. [Modes for carrying out the invention]

[0038] The present invention will be described with reference to the drawings with respect to specific embodiments, but will not be limited thereto, and will be limited only by the claims. The drawings provided are schematic and non-limiting. In the drawings, the size of elements may be exaggerated or not drawn to their actual dimensions for illustrative purposes. Where the term “equipped with” is used in this specification and claims, other elements or processes are not excluded. Where an indefinite article such as “a” or “an” is used when referring to a singular noun, or a definite article such as “the,” this includes the plural noun unless the opposite is specifically stated.

[0039] The term “equipped with” as used in a claim should not be interpreted as being limited to the means listed thereafter. The term “equipped with” as used in a claim does not exclude other elements or processes. Therefore, the expression “the device is equipped with means A and B” should not be limited to a device consisting solely of elements A and B. With respect to the present invention, this expression merely means that the only relevant components of the device are A and B.

[0040] Furthermore, in this specification and its claims, terms such as "first," "second," "third," etc., are used to distinguish similar elements and are not necessarily used to describe a particular order or chronological sequence. It should be noted that the terms used in this manner are interchangeable under appropriate circumstances, and that embodiments of the inventions described herein may operate in an order other than that described or explained herein.

[0041] Figure 1 schematically shows a luggage locker bank 50. The luggage locker bank 50 comprises a number of individually secure luggage lockers. Each luggage locker is indicated in the drawing by a number between 1 and 20. Each luggage locker has one of three sizes. Each size is indicated in Figure 2 by the following reference numerals: S, which refers to the smallest size of the luggage locker; L, which refers to the largest size of the luggage locker; and M, which refers to an intermediate or medium size between S and L.

[0042] The luggage locker bank is configured to allow luggage to be stored in one of its multiple luggage lockers for a maximum number of days equal to a first integer. This first integer is denoted as N below, for example, 4 days (N=4). If a luggage is still in the luggage locker after the above maximum number of days, it will be removed by the delivery person the following day. In this way, luggage lockers will not get stuck due to unclaimed luggage.

[0043] The luggage locker bank 50 is equipped with a locker bank computer 51, which includes a processor and memory and is configured to obtain data related to the luggage locker bank 50 and transmit the data to the luggage locker bank management system 30. Each of the individually secure luggage lockers 1 to 20 is equipped with a door and a locking mechanism.

[0044] Figure 2 is a flowchart of a computer implementation method according to an embodiment of the present invention for predicting the daily available capacity of a luggage locker bank.

[0045] The method is performed for a number of lockers equal to a second integer, and for a specific day. The second integer is denoted as P below. The specific day is denoted as j below, and corresponds to a day on which a user of the lockers, such as a delivery person, wishes to deliver at least one package to the above locker bank. Each package deposited in the above locker bank 50 is permitted to be deposited for a maximum number of days equal to N from the day the package is deposited in the locker. Thus, a package can be deposited in a locker from a deposit date such as the above specific future day j, and can be deposited for a maximum of the day j+N-1, where j+N-1 is the latest permitted deposit date.

[0046] As illustrated, the above method comprises the following prediction steps: In the first step 100, for each day between the above-mentioned specific future day j and the above-mentioned latest acceptable storage day j+N-1, and for each day between day j and a number of days equal to N-1 prior to this future day, the number of packages delivered or scheduled to be delivered to the package locker bank 50 is received. A number of days equal to N-1 prior to the above future day means going back to day j-N+1.

[0047] In the second step 110, a third number Sj is estimated. This third number Sj corresponds to the number of package lockers that will contain packages on the specific future day j mentioned above. This estimate is based on the number of packages that have been delivered or are scheduled to be delivered, and the package retrieval rate from the package lockers. The retrieval rate depends on the number of days that packages still in the package lockers have been stored up to that point.

[0048] The third number Sj is given by the following formula:

number

[0049] Here, Del x This corresponds to the number of packages delivered to the package locker bank on day x. PckRate y,zThis corresponds to the pickup rate on day y for packages delivered on day z. Day y is later than day z. x This corresponds to deliveries that have already been made or scheduled for delivery and have been notified by the delivery person for day x.

[0050] When a baggage locker bank is about to enter production of a method for predicting its daily available capacity, the retrieval rate applied to the baggage locker bank is based on a set of training data on retrieval rates previously observed at the same or similar baggage locker bank. The method is initiated using the retrieval rate derived from the training data set when the baggage locker bank enters production of a method for predicting its daily available capacity. Once the baggage locker bank is in operation, the retrieval rate is updated daily using daily data on the flow of packages associated with the baggage locker bank, as described in the following paragraphs.

[0051] To train the system regarding the retrieval rate, multiple parcel lockers are classified into pairs (locker, item), i.e., the type of item stored in a given parcel locker, and the following two variables, namely the total number of deliveries and the seniority, can be used for classification. Therefore, the applicable model will differ depending on the number of parcels delivered to the locker and the seniority of the parcel locker.

[0052] For example, if a pair (locker, item) has a total number of deliveries that is strictly greater than 365, a separate model applies. Otherwise, since adjacent lockers with the same partner are considered nearly identical, a different model applies to treat the nearest individual items as a single unit.

[0053] Two pairs (lockers, items) with identical models (not separate models) have the same training dataset. Therefore, they have the same retrieval rate.

[0054] Each pickup rate is estimated daily by regression outside of the luggage locker bank's operating hours. The regression considers known actual pickup rates for several past days, taking into account whether or not those days include holidays.

[0055] In the third step 120, a fourth number Dj is calculated. The fourth number Dj corresponds to the number of available parcel lockers in the parcel locker bank on the specific future day j. The fourth number Dj is obtained by subtracting the third number from the second integer, i.e., Dj = P - Sj. Therefore, the third number corresponds to the number of available parcel lockers remaining on the day being calculated. In other words, the third number for day j corresponds to the number of available parcel lockers for day j considering only the parcels currently stored or announced to be delivered into a parcel locker for day j, at a specific time such as 6 a.m.

[0056] In this embodiment, the method comprises a fourth step 130, which is a step to reduce the risk of error regarding the availability of luggage lockers on a desired day. The fourth step 130 comprises comparing a fourth number with a threshold and indicating that there are no available luggage lockers if the fourth number is smaller than the threshold.

[0057] Prediction steps 100-130 are repeated for each size of luggage locker, S, M, and L. In each iteration, the second integer P corresponds to the number of luggage lockers of the corresponding size, the number of deliveries to the luggage locker bank corresponds to that size, and if there are differences in the retrieval rate, the retrieval rate corresponds to that size.

[0058] In the embodiment of the luggage locker bank shown in Figure 1, the luggage locker bank 50 includes a first group of luggage lockers, shown in light gray and assigned only to the first user; a second group of luggage lockers, shown with diagonal lines in Figure 1 and assigned only to the second user; and a third group of luggage lockers, shown in white and shared by the first and second users.

[0059] In this configuration, prediction steps 110-130 are repeated for each group of luggage lockers. For each group of luggage lockers, the second integer P corresponds to the number of luggage lockers in that group.

[0060] Optionally, the method may also consider whether the luggage locker bank has any broken lockers. In this example, the second integer P is recalculated taking into account the absence of broken lockers.

[0061] In another embodiment of the method shown in Figure 2, the method may include an additional step between the third step 120 and the fourth step 130 described above.

[0062] In additional step 125, a fourth number is calculated for each day between day j and day j+N-1. The fourth number corresponds to the number of available luggage lockers for each day of this period. The fourth number is based on the third number and the number of packages received that have been delivered or are scheduled to be delivered and notified by the delivery person to the above-mentioned specific type of luggage lockers in the luggage locker bank on the day subject to calculation. Thus, the fourth number corresponds to the number of available luggage lockers of the above-mentioned specific type, taking into account all packages already deposited and all packages scheduled to be deposited, for each day of the N days during which a package can be deposited if it is deposited on day j, i.e., the period during which a package can be deposited in the luggage lockers of the luggage locker bank.

[0063] The fourth number Ej is given by the following formula: Regarding the above desired date,

number

number

[0064] Here, k is a computation parameter, and it is an integer because the computation applies to each day.

[0065] In the above embodiment, the fourth step 130 comprises comparing the above fourth number Ej with a threshold τ for each day from the above desired day j to day j+N-1, which is the last day on which luggage can be deposited. If each of the compared fourth numbers is greater than the above threshold, a signal is given indicating that a luggage locker is available for the luggage in the luggage locker bank 50 in question. In other words,

number

[0066] Figure 3 is a flowchart of a computer implementation method for determining the destination of one or more packages to the appropriate luggage locker bank 50 in a luggage locker bank network.

[0067] In the first step 300 of the method, a request is received to deposit at least one piece of luggage on a specific future day j in one of several luggage locker banks located near the destination address.

[0068] In the second step 310, the luggage locker bank closest to the destination address is selected. Then, in the third step 320, by performing the prediction steps 100-120 of the computer implementation method shown in Figure 2, it is determined whether there is a suitable luggage locker available for the luggage on the specified future day j.

[0069] A luggage locker is considered suitable for storing luggage if it is large enough to accommodate luggage and is not exclusively assigned to a user other than the user requesting it. Steps 310 and 320 may be performed in the same process because the present invention makes it possible to identify a luggage locker bank that has available lockers.

[0070] In step 320, if there is no suitable luggage locker available for the above luggage in the nearest luggage locker bank selected in step 310, steps 310 and 320 are repeated as many times as necessary until a suitable luggage locker is found available on the above specific future date. However, each iteration is performed in a different luggage locker bank, and the luggage locker bank selected each time is the next closest luggage locker bank to the destination address after the previously selected luggage locker bank.

[0071] Other priorities may be set between different features. For example, accessible areas may be classified first by the size of the luggage locker, then by the allocated area (luggage lockers assigned to a user first, then luggage lockers shared among different users), and then by comfort zones.

[0072] If a suitable luggage locker is found in step 320, the method in step 4, step 330, sends a message indicating the selected luggage locker bank that has a suitable luggage locker available for depositing the above luggage on the above specific future day j, and asks whether or not to reserve the above suitable luggage locker.

[0073] In the methods shown in Figures 2 and 3, estimations and calculations are performed on a daily basis. That is, all pickup rates are given for each day, and all figures related to delivery or pickup are given for each day.

[0074] It appears that during the delivery phase, which begins at the start of the day and typically occurs between 7:00 AM and 4:00 PM, new package lockers are not (or are only available in limited quantities). Package lockers from which packages were delivered on that day or a previous day are not available until 5:00 PM. Therefore, daily modeling can be performed before the delivery phase, for example, at 6:00 AM.

[0075] Figure 4 is a schematic diagram of the luggage locker bank management system 30. The luggage locker bank management system 30 comprises at least one computerized luggage locker bank 50 as defined in Figure 1. The luggage locker bank management system comprises a central unit 40. The central unit 40 comprises a prediction unit 60, a destination determination unit 70, and a data storage unit 80. The central unit 40 and its units 60, 70, and 80 communicate with the locker bank computers of the luggage locker banks 50. The data storage unit 80 is configured to receive and store all information transmitted by at least one computerized luggage locker bank 50.

[0076] The prediction unit 60 is configured to perform a computer implementation method for assigning one or more packages to appropriate luggage locker banks 50 in the luggage locker bank network shown in Figure 3. For this purpose, the prediction unit 60 comprises a receiving module 61, a first computing module 62, a second computing module 63, and a regression module 64.

[0077] The receiving module 61 is configured to receive from the data storage unit 80 the number of packages delivered or scheduled to be delivered to the package locker bank for each day between a specific future day j and a future day N-1 days after that specific day j. The specific day corresponds to a day on which the user wishes to have at least one package delivered to the package locker bank. In other embodiments, the receiving module 61 is composed of a plurality of separate modules, each module being specialized to receive a particular type of information.

[0078] The first computation module 62 is configured to estimate a third number Sj. The second computation module 63 is configured to compute a fourth number Dj.

[0079] The first calculation module 62 comprises a regression module 64. The regression module 64 is configured to estimate each retrieval rate by regression every day, outside of the operating hours of the luggage locker bank. The regression considers the actual retrieval rates for several past days that are known from information obtained from actual retrieval dates for luggage deposited in the past, and takes into account whether or not holidays are included in the above several days.

[0080] The destination determination unit 70 is configured to determine the destination of each requested piece of luggage to be placed in the appropriate luggage locker bank 50. The destination determination unit 70 comprises a request module 71, a selection module 72, and a transmission module 73.

[0081] The request module 71 is configured to receive requests to deposit luggage on a specific future day j in one of several luggage locker banks 50 located near the destination address.

[0082] The selection module 72 is configured to select the luggage locker bank closest to the destination address and to work in cooperation with other modules of the prediction unit 60 to determine whether there is a suitable luggage locker available for the luggage on the specified future day j.

[0083] The transmission module 73 is configured to transmit information indicating the nearest luggage locker bank that has a suitable luggage locker available for depositing the above luggage on the above-mentioned specific future day j, and to ask the user, i.e., the delivery person, whether or not they should reserve the selected luggage locker.

[0084] The present invention provides a computerized implementation method and a computerized system that helps maximize the efficiency of a package locker bank by predicting the daily available capacity of the package locker bank and minimizing the risk of undelivered packages.

Claims

1. A computer-implemented method for predicting the daily available capacity of a luggage locker bank (50), wherein the locker bank (50) comprises a plurality of individually securable luggage lockers (1-20), and allowing a luggage to be deposited in one of the plurality of luggage lockers (1-20) of the luggage locker bank (50) for a maximum number of days equal to a first integer hereinafter denoted as N, the method comprising, for a number of luggage lockers equal to a second integer (P), the following plurality of prediction steps: - Receiving (100) the number of delivered or scheduled-to-be-delivered luggages at the luggage locker bank for each day between a future day (j) and the day N-1 days before the selected future day (j), wherein the selected day (j) corresponds to the day on which a user wishes to deliver at least one luggage to the luggage locker bank, - Estimating (110) a third number (Sj) corresponding to the number of luggage lockers that will contain luggages on the selected future day (j), wherein the estimation is based on the received number of delivered or scheduled-to-be-delivered luggages and the withdrawal rate of luggages from the luggage lockers, and the withdrawal rate is given for each day and depends on the number of days for which each luggage still deposited in the luggage locker has been deposited so far, - Calculating (120) a fourth number (Dj) corresponding to the number of available luggage lockers in the luggage locker bank on the selected future day (j), wherein the fourth number (Dj) is obtained by subtracting the third number (Sj) from the second integer (P), A computer-implemented method comprising the above steps.

2. The computer-implemented method according to claim 1, wherein each withdrawal rate is estimated by regression outside the business hours of the luggage locker bank every day, and in the regression, the known actual withdrawal rates for a plurality of past days are taken into account, and whether the plurality of days include holidays is taken into account.

3. The computer-implemented method according to claim 1 or 2, wherein the plurality of prediction steps further comprise a confirmation step after the calculation of the fourth number, and the confirmation step comprises comparing the fourth number with a threshold value and indicating that there are no available luggage lockers when the fourth number is smaller than the threshold value.

4. After the calculation step, a second calculation for each day from the future day (j) to the last day (j + N - 1) on which a package can be deposited, and a confirmation step are provided. The fourth number corresponds to the number of available package lockers for each day in this period, and the fourth number is based on the third number and the received number of packages that were delivered or scheduled for delivery and notified by the deliverer to the specific type of package locker in the package locker bank on the day targeted for the calculation. The confirmation step includes comparing each calculated fourth number with a threshold value (τ), and indicating that there are no available package lockers when each calculated fourth number is less than the threshold value. The computer-implemented method according to claim 1 or 2.

5. The package locker bank includes a first class of package lockers of a first size and at least one second class of package lockers of at least one second size larger than the first size. The plurality of prediction steps are repeated for each class of package lockers, and in each class of package lockers, the second integer (P) corresponds to the number of package lockers of the corresponding class of package lockers. The computer-implemented method according to any one of claims 1 to 4.

6. The package locker bank has a first group of package lockers assigned only to a first user and a second group of package lockers assigned to the first user and at least one other user. The plurality of prediction steps are repeated for each group of package lockers, and for each group of package lockers, the second integer (P) corresponds to the number of package lockers of the corresponding group. The computer-implemented method according to any one of claims 1 to 4.

7. The luggage locker bank has a first group of luggage lockers assigned only to a first user and a second group of luggage lockers assigned to the first user and at least one other user, and each group includes a first group of luggage lockers of a first size and at least one second group of luggage lockers of a second size larger than the first size. The plurality of prediction steps are repeated for each group of luggage lockers, and for each group of luggage lockers, the second integer (P) corresponds to the number of luggage lockers in the corresponding group for the corresponding group. The computer-implemented method according to any one of claims 1 to 4.

8. When the luggage locker bank includes a malfunctioning luggage locker, the second integer (P) is recalculated taking into account that the number of malfunctioning luggage lockers is not included. The computer-implemented method according to any one of claims 1 to 7.

9. A computer-implemented method for determining the destination of one or more pieces of luggage to an appropriate luggage locker bank, the method comprising: - Receiving a request (300) to store at least one piece of luggage on a future day (j) in one of a plurality of luggage locker banks located near the destination address. - Selecting (310) the luggage locker bank closest to the destination address and determining whether there is an appropriate luggage locker available for the luggage on the selected future day by executing the computer-implemented method according to any one of claims 1 to 7 for predicting the daily available capacity of the luggage locker bank. (320), - If there is no appropriate luggage locker available for the luggage in the closest luggage locker bank, repeating the selection and prediction steps the required number of times until an appropriate luggage locker available on the selected future day (j) is found. Here, the luggage locker bank in each iteration is the luggage locker bank closest to the destination address next to the previously selected luggage locker bank. - Transmitting (330) an indication of the luggage locker bank having an appropriate luggage locker available for storing the luggage on the selected future day (j) and querying the user whether the available appropriate luggage locker should be reserved. A computer-implemented method comprising.

10. A luggage locker bank management system (30) comprising at least one computerized luggage locker bank (50), wherein the at least one computerized luggage locker bank (50) comprises a plurality of individually securable luggage lockers (1 - 20), permits storing luggage in one of the plurality of luggage lockers of the at least one computerized luggage locker bank (50) for a maximum number of days equal to a first integer (N), the at least one computerized luggage locker bank (50) further comprises at least one locker bank computer (51), the at least one locker bank computer (51) comprises at least one processor and a memory, is configured to obtain data related to the at least one computerized luggage locker bank (50) and transmit the data outside the at least one computerized luggage locker bank (50), the number of luggage lockers is equal to a second integer (P), each of the plurality of individually securable luggage lockers comprises at least one door and at least one locking mechanism, and the luggage locker bank management system (30) comprises a prediction unit (60) communicating with the locker bank computer (51) of the at least one computerized luggage locker bank (50), and the prediction unit (60) is: - A receiving module (61) configured to receive, for each day between a future day (j) and a day N - 1 days before the selected future day (j), the number of delivered or to-be-delivered luggage in the luggage locker bank, wherein the selected future day (j) corresponds to the day on which a user wishes to deliver at least one piece of luggage to the luggage locker bank. - A first calculation module (62) configured to estimate a third number (Sj) corresponding to the number of luggage lockers that will contain luggage on the selected future day (j), wherein the estimation is based on the received number of delivered or to-be-delivered luggage and the rate of luggage retrieval from the luggage lockers, and the retrieval rate depends on the number of days each piece of luggage has been stored in the luggage locker. - A second calculation module (63) configured to calculate a fourth number (Dj) corresponding to the number of available luggage lockers in the luggage locker bank on the selected future day (j), where the fourth number (Dj) is obtained by subtracting the third number (Sj) from the second integer (P). A luggage locker bank management system (30) comprising the above. **Claim 11** The first calculation module (62) comprises a regression module (64), and the regression module (64) is configured to estimate each pick-up rate by regression outside the business hours of the luggage locker bank every day. In this regression, the actual pick-up rates for a plurality of past days known from information obtained from the actual pick-up dates of the luggage deposited in the past are considered, and whether or not the plurality of days include holidays is considered. The luggage locker bank management system (30) according to claim 10. **Claim 12** The second calculation module (63) comprises a confirmation module, and the confirmation module is configured to compare the fourth number with a threshold value and indicate that there are no available luggage lockers when the fourth number is smaller than the threshold value. The luggage locker bank management system (30) according to claim 10 or 11. **Claim 13** The at least one computerized luggage locker bank (50) has a first group of luggage lockers assigned only to a first user and a second group of luggage lockers assigned to the first user and at least one other user. Each group comprises a first group of luggage lockers of a first size and at least one second group of luggage lockers of a second size larger than the first size. The plurality of prediction steps are repeated for each group of each group of luggage lockers. For each group of luggage lockers, the second integer (P) corresponds to the number of luggage lockers in the corresponding group of the corresponding group. The luggage locker bank management system (30) according to any one of claims 10 to 12. **Claim 14** When the at least one computerized luggage locker bank (50) comprises a malfunctioning luggage locker, the second integer (P) is recalculated taking into account that the number of malfunctioning luggage lockers is not included. The luggage locker bank management system (30) according to any one of claims 10 to 13. **Claim 15** It further includes a destination determination unit (70) configured to determine the destinations of one or more pieces of luggage to an appropriate luggage locker bank (50), and the destination determination unit (70) is: - A request module (71) configured to receive a request for storing at least one piece of luggage in one of the plurality of luggage locker banks located near the destination address on a selected future day (j). - Select the luggage locker bank closest to the destination address, and cooperate with the prediction unit (60) to identify whether there is an appropriate luggage locker available for the luggage on the selected future day. And if there is no appropriate luggage locker available for the luggage in the closest luggage locker bank, it is configured to repeat the process the required number of times until an appropriate luggage locker available on the selected future day (j) is found. Here, the luggage locker bank in each iteration is the luggage locker bank closest to the destination address next to the previously selected luggage locker bank. - A transmission module (73) configured to transmit an indication of the luggage locker bank having an appropriate luggage locker available for storing the luggage on the selected future day (j), and to inquire whether the available appropriate luggage locker should be reserved. The luggage locker bank management system (30) according to any one of claims 10 to 14, comprising the above components.

Images