Field of the Invention

[1] This invention relates generally to order fulfilment, sorting and dispatch system. More particularly, this invention relates to an intelligent demand predictive pre-emptive pre-sorting e- commerce order fulfilment, sorting and dispatch system for dispatch routing optimisation.

Background of the Invention

[2] Figure 1 shows an industrial sorting centre according to the prior art. Such industrial sorting centres 108 sort parcels on a massive scale such as in excess of 100,000 parcels per day and may be utilised by operators such as Australia Post, Amazon and the like.

[3] Specifically, incoming parcels 101 received that various depots 102 are places in inbound parcel queues 103 of the sorting centre 108. The parcels are then scanned 104 such as by utilising various scanning techniques including OCR and the like so as to allow for substantially automated mechanised conveyor sorting 105.

[4] The mechanised conveyor sorting 105 physically allocates the parcels to various sorting sort bins 106. The sorting sort bins are arranged according to geographic zones. As such, once the parcels have been allocated in this manner to various sorting sort bins, parcels may be routed 107 therefrom by being allocated to and handled by geographically respective depots for dispatch.

[5] Given the massive scale of past delivery, fixed geographic zoning of the sorting sort bins is appropriate given that zonal volume fluctuation is substantially invariant from day-to-day.

[6] As such, the routing efficiency of such industrial sorting centres 108 is not substantially affected by such a small zonal volume fluctuation.

[7] However, we discovered that routing efficiency of systems for fulfilment and dispatch of smaller amounts of parcels may be substantially degraded by time period zonal volume fluctuation.

[8] For example, for fulfilment and dispatch systems, as may be typical for e-commerce platform providers fulfilling less than 1000 parcels per day for example, we discovered that day-to-day zonal volume fluctuation substantially degraded the routing efficiency thereof and drove up delivery costs.

[9] As such, we devised a fulfilment, sorting and dispatch system to overcome or substantially ameliorate at least some of the deficiencies of the prior art, or to at least provide an alternative.

[10] It is to be understood that, if any prior art information is referred to herein, such reference does not constitute an admission that the information forms part of the common general knowledge in the art, in Australia or any other country. Summary of the Disclosure

[11] As is disclosed herein, we have devised an intelligent demand predictive pre-emptive presorting e-commerce order fulfilment and dispatch system configured for dispatch routing optimisation by intelligently predicting order demand for a future time period.

[12] By being able to predict order demand for various geographic regions for a future time period, a delivery geographic region may be dynamically zoned according to predicted demand.

[13] As such, sort bins may be preconfigured according to the expected demand per region so as to allow for packaging and placement of order parcels therein prior routing.

[14] It should be noted that the predictive sort bin placement may be conducted before receipt of e-commerce orders wherein such subsequent e-commerce orders have been intelligently anticipated by the system. For example, sort bins may be predictively placed prior 10 PM the day before delivery wherein further e-commerce orders may be received overnight and packaged and placed in the predictively pre-placed sort bins for dispatch by 8 AM the following day, such a process enabling the optimisation of routing dispatch.

[15] As such, with the foregoing in mind, in accordance with one embodiment, there is provided an intelligent demand predictive pre-emptive pre-sorting e-commerce order fulfilment, sorting and dispatch system for dispatch routing optimisation, the system comprising: an e-commerce shopping cart module configured for receiving orders from a plurality of client terminals across a data network; a demand prediction controller operably coupled to an order history database, the demand prediction controller configured for predicting order demand for a future time period using historical order data selected from the order history database, the predicted order demand comprising at least order volume correlated to a plurality of geographic regions; a dynamic zoning controller configured for dynamically zoning a geographic region into a plurality of zones according to the predicted order demand; a sort bin configuration and placement controller configured for sort bin allocation and placement according to the plurality of zones; and a fulfilment controller operably coupled to an inventory database, the fulfilment controller configured for fulfilling the orders using the inventory database according to the sort bin allocation and placement.

[16] Sort bin allocation and placement may occur before receiving at least a subset of the orders.

[17] The future time period may represent the next day.

[18] The dynamic prediction controller may comprise a machine learning module trained utilising training data comprising the historical order data, the machine learning module configured for generating optimising parameters for optimising a trained machine, the trained machine configured for outputting the predicted order demand for the future time period.

[19] The trained machine may be an artificial neural network (ANN). [20] The historical order data may comprise data fields comprising at least one of delivery date, goods ID, number of goods per goods ID, delivery address, price, goods volume, goods weight, time of order and date of order.

[21] The historical order data may comprise customer data may comprise at least one of customer ID, customer address, customer demographics including age, gender and income and the like

[22] The demand prediction controller may be further configured for predicting goods type for at least a subset of the predicted order demand.

[23] A predicted order parcel combined volume may be calculated by the system according to the goods type prediction.

[24] At least one of the dynamic zoning controller and the sort bin configuration and placement controller has as input the predicted order parcel combined volume.

[25] The fulfilment controller may be configured for allocating each order to a sort bin ID.

[26] The fulfilment controller may be configured for printing a dispatch label comprising the sort bin ID.

[27] Other aspects of the invention are also disclosed. Brief Description of the Drawings

[28] Notwithstanding any other forms which may fall within the scope of the present invention, preferred embodiments of the disclosure will now be described, by way of example only, with reference to the accompanying drawings in which:

[29] Figure 1 shows an industrial sized sorting centre of the prior art;

[30] Figure 2 shows an intelligent demand predictive pre-emptive pre-sorting e-commerce order fulfilment sorting and dispatch system for dispatch routing optimisation in accordance with an embodiment;

[31] Figure 3 illustrates predictive dynamic zoning in accordance with an embodiment; and

[32] Figure 4 illustrates an artificial intelligence implementation of a demand prediction controller in accordance with an embodiment.

Description of Embodiments

[33] Figure 2 shows a parcel ordering and delivery network 200. The network 200 comprises an intelligent e-commerce order fulfilment sorting and dispatch system 203.

[34] In embodiments, the system 203 may take the form of a standalone or interconnected system of computing devices comprising processors for processing digital data and memory devices for storing digital data including computer program code instructions, the memory devices being operably coupled to the processors. As such, each processor is configured for retrieving such computer program code instructions from such memory devices for interpretation and execution in the implementation of the computational processing tasks described herein. In this regard, such computer program code instruction modules and associated hardware is the case may be may be logically divided into the various controllers described herein.

[35] The system 203 comprises an e-commerce shopping cart module 204 configured for receiving order data from a plurality of client terminals 201 across a data network 202, such as the Internet.

[36] Specifically, the system 203 may employ a web interface to expose an e-commerce web resource comprising a plurality of goods which may be ordered online by customers utilising the client terminal 201. The e-commerce web resource may be operably coupled with an inventory database

206 so as to host the various available goods via the resource for e-commerce purchasing.

[37] Now, the system 203 is characterised in being able to intelligently predict order demand so as to allow for dynamic predictive geographic region zoning for routing dispatch optimisation.

[38] Specifically, as is shown in figure 2, the system 203 comprises a demand prediction controller

207. The demand prediction controller 207 may utilise historical order data from an order history database 208 to predict order demand for a future time period.

[39] Specifically, the demand prediction controller 207 may be configured for predicting order volumes according to geographic region. In further embodiments, the demand prediction controller may be configured for predicting other relevant parameters for routing optimisation including goods type (affecting delivery volumes) and the like.

[40] Having predicted the future order demand volume correlated with geographic region, a dynamic zoning controller 209 is able to dynamically allocate geographic zones according to the predicted demand.

[41] Once the zones have been dynamically configured utilising the dynamic zoning controller 209, sort bin configuration controller 210 may be configured for dynamically allocating sort bins according to the dynamically generated zones.

[42] Thereafter, the system 203 may comprise a fulfilment controller 205 configured for fulfilling e-commerce orders. Specifically, for the orders received from the client terminals 201, the fulfilment controller 205 is able to fulfil the orders including allocating the ordered goods to the dynamically configured sort bins.

[43] In this regard, the fulfilment controller 205, when fulfilling items, may print out delivery labels using a connected printer device to be adhered to each parcel and wherein each delivery label designates the relevant sort bin for placement. In embodiments, a bin placement/loading mechanism 212 may be employed for the automated placement of the parcels within the allocated sort bins. [44] Such mechanism 212 may comprise one or automatically controlled conveyor belts that conveyed stock from the stock warehouse 213 into the appropriate sort bins 212. The mechanism 212 may comprise electronic scanning devices which read computer readable data of each parcel prior placement and once loaded into at least one of the sort bins 211 prior dispatch.

[45] As such, stock from the stock warehouse 213 is allocated to and placed in the relevant various dynamically allocated sort bins 211.

[46] Once the parcels have been placed in the dynamically allocating sort bins 211, routing 214 may then commence wherein the routing optimisation is aided by the parcels having been intelligently allocated to dynamically configured sort bins 214 correlating to predicted volume and geographic region.

[47] Figure 3 illustrates the predictive dynamic zoning implemented by the system 203.

[48] Specifically, there is shown a first geographic zoning 301 wherein a geographic region has been divided into 8 zones numbered Zl - Z8. As alluded to above, for industrial sized sorting centres, such zones may be typically static.

[49] However, utilising the present system 203, the system 203 is able to dynamically change the zoning allocation so as to optimised grouting delivery.

[50] Specifically, the first geographic zoning 301 may represent zoning at a first time period and second geographic zoning 302 may represent zoning and a second time period wherein the zoning at the second time period has been predictively dynamically zoned by the system 203.

[51] For example, for the first zonal allocation 301, zone Zl may be allocated to the Parramatta region, zone Z8 allocated to Sutherland shire, zone Z2 to the lower North Shore, zones Z3 - Z6 to the eastern suburbs and zone Z7 to the inner West.

[52] However, for a future time period, the demand prediction controller 207 may predict an increased order demand for the lower North Shore. As such, when performing the predictive dynamic zoning, the dynamic zoning controller 209 may allocate zone Z2 - Z5 to the lower North Shore and only 2 zones Z6 - Z7 to the eastern suburbs in anticipation of the predicted expected increase of order vol ume for the lower North Shore.

[53] As such, at the future time, the greater number of order parcels are allocated to sort bins Z2 - Z5 so as to allow for routing optimisation including in utilisation of 4 couriers as opposed to previously one for the lower North Shore.

[54] As alluded to above, the dynamic zoning and sort bin configuration and placement may be performed in advance of receiving orders wherein, for example, dynamic zoning and sort bin configuration and placement may be completed by 10 PM at night so as to allow for adequate time for layout and placement of the sort bins 211 wherein further e-commerce orders may be received overnight which may also be packaged and placed in the pre-configured sorting sort bins 211.

[55] Figure 400 illustrates an artificial intelligence implementation 400 of the demand prediction controller in accordance with an embodiment.

[56] As can be seen from the implementation 400, the demand prediction controller 207 has as input historical order data 401.

[57] In an embodiment, the demand prediction controller 207 utilises artificial intelligence for the conversion of the historical order data to predicted demand 407.

[58] Specifically, in the embodiment shown, the demand prediction controller 207 may utilise a machine learning module 403 which trains utilising training data 402 (being the historical order data

401) to generate optimising parameters 404 to optimise the trained machine 405. As such, the trained machine 405 is able to output a predicted demand for 107 for a future time period.

[59] In embodiment, the trained machine 405 may take the form of an artificial neural network

(ANN).

[60] The predicted demand 407 comprises at least order volume and geographic region.

[61] Various data of the order data 401 may be utilised for prediction of the order demand prediction 407. For example, the artificial intelligence implementation of the demand prediction controller 207 may take into account time period parameters such as days of the week, seasonal variability and the like. Other factors may include order trends including those affected by external factors such as the media, weather patterns and the like.

[62] As can be appreciated, the utilisation of artificial intelligence for the demand prediction controller 207 embodiments allows for the demand prediction controller 207 to tease out potentially unintuitive nuances from the input data affecting demand beyond human being comprehension.

[63] In embodiments, the historical order data may comprise data fields comprising delivery date, goods ID, number of goods per goods ID, delivery address, price, goods volume, goods weight, time of order, date of order and the like. The historical order data may further comprise customer data including customer ID, customer address, customer demographics including age, gender and income and the like.

[64] As such, having predicted the order demand for various geographic regions, dynamic zoning 408 may be performed corresponding to the order volume and region. For example, for geographic regions for which high order demand is predicted, a greater number of zones may be allocated.

[65] Thereafter, bin placement configuration 409 occurs wherein sort bins 211 are allocated and placed according to the zones. For example, for the 8 zones of figure 3, five sort bins may be utilised for each zone such that the total bin allocation represents 40 sort bins 211. [66] During bin placement configuration 409 and, as alluded to above, even after the bin placement configuration 409, orders may be received 410 via the e-commerce shopping cart module 204 which, when fulfilled utilising the fulfilment controller 205 are packaged and physically placed 411 from stock 412 in the relevant allocated sorting sort bins 211.

[67] Then, later, at the close of the order fulfilment period, routing 413 occurs wherein couriers collect the allocated packages from the allocated sort bins for routing and dispatch. Specifically, utilising the example provided above with reference to figure 3, four couriers may be allocated for each of zones Z2 - Z5 for the second zonal configuration 302 as opposed to the single courier for the previous first zonal configuration 301 so as to increase the number of couriers correspondingly to the increased order volume.

[68] Having collected the parcels, routing may occur wherein the delivery route for each courier is calculated and optimised.

[69] As alluded to above, it should be noted that the predicted order volume may include both the number of parcels and also the volume occupied by the parcels. For example, the demand prediction controller may also predict the type of goods that are going to be ordered so as to be able to calculate a combined volume of the predicted number of goods such that the goods may be allocated to sort bins for couriers having capacity for such volume.

[70] The foregoing description, for purposes of explanation, used specific nomenclature to provide a thorough understanding of the invention. However, it will be apparent to one skilled in the art that specific details are not required in order to practice the invention. Thus, the foregoing descriptions of specific embodiments of the invention are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed; obviously, many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, they thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. It is intended that the following claims and their equivalents define the scope of the invention.