Online order batching method and system based on order prediction reservation mechanism

Abstract

The present invention proposes an online order batching method based on an order prediction reservation mechanism, which includes the following steps: S1. Based on a machine learning method, predict future orders according to lane similarity, and use order data to train an order prediction scoring model; S2. Score the orders in the order pool based on the order prediction scoring model, design an order reservation algorithm to decide whether to retain the order in the order pool or release it to the batching pool; S3. According to the number of idle pickers in the system, use a heuristic algorithm to batch, sort, allocate, and route the orders in the batching pool. The algorithm proposed by the present invention is applicable to a person-to-goods warehousing system where orders arrive online, optimizes the batching result, thereby reducing the picking labor cost, improving the order turnover rate, and enhancing the service level.

Description

Technical Field

[0001] The present invention relates to the fields of warehousing systems and order picking, and particularly to an online order batching method and system based on an order prediction reservation mechanism. Background Art

[0002] With the development of e-commerce, hundreds of thousands of orders every day pose great challenges to the timeliness and operation cost of the warehousing picking system. Traditional e-commerce warehouses generally adopt static batch picking, that is, all orders arriving every day are divided into several waves, and the orders are batch-picked in each wave, and the next wave is executed after the current wave is completed. This method is easy to handle, but the disadvantage is that it deviates from the actual problem, and there is room for optimization in terms of cost and timeliness. In dynamic batch picking, orders arrive at the system over time, and batches are continuously generated after the orders are available.

[0003] Currently, according to the different time points of updating the picking list, the dynamic picking list strategy can be divided into two types: one is to update the picking list at the picking table, and the other is to update the picking list during the picking process. The disadvantage of the former is that batches are generated according to the current order pool information, and once a batch is assigned, it cannot be updated. It does not consider the impact of future arriving orders. It is possible that subsequent arriving orders can form a better batch, ignoring the global optimum; while the latter is difficult to implement in operation because batches can also be changed during the picking process.

[0004] There is a lack of an online order batching algorithm in the prior art that applies order big data information and considers the situation of future arriving orders.

[0005] It should be noted that the information disclosed in the above background art section is only used for understanding the background of the present application, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to overcome the disadvantages of ignoring the global optimum when updating the picking list at the picking table and being difficult to implement in operation when updating the picking list during the picking process in the above background art, and provide an online order batching method and system based on an order prediction reservation mechanism.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions:

[0008] An online order batching method based on an order prediction reservation mechanism includes:

[0009] S1. Based on a machine learning method, predict future orders according to the roadway similarity, and use order data to train an order prediction scoring model;

[0010] S2. Score the orders in the order pool based on the order prediction scoring model, design an order retention algorithm to determine whether to retain the orders in the order pool or release them to the batch pool;

[0011] S3. According to the number of idle pickers in the system, use a heuristic algorithm to batch, sort, allocate, and route the orders in the batch pool.

[0012] In some embodiments, step S1 includes:

[0013] S11. Provide order data, which is the training data of the model and is the order information in the warehousing system, including: order arrival time, items in the order, item locations, and item quantities;

[0014] S12. Construct the features of the data based on the order data; the features include: the order arrival time, which is a continuous value; the roadway location of the goods contained in the order, which is a 0-1 value. 0 indicates that there are no items to be picked in the roadway for this order, and 1 indicates that there are items to be picked in this roadway for this order; the quantity of items in the order;

[0015] S13. Construct the data labels based on the order data; the label is defined as: the sum of the similarities between the current order and the orders arriving in a future period of time;

[0016] S14. Divide the data with constructed features and labeled data into a training set and a test set, select a suitable machine learning regression model, including but not limited to: tree models, ensemble models, deep learning models; select a suitable model evaluation criterion, such as mean squared error; train and validate the model on the training set and evaluate the model on the test set.

[0017] In some embodiments, the definition of the order similarity is: |lane i ∩ lane j| / |lane i ∪ lane j|, that is, the number of lanes that both orders need to enter divided by the number of non-repeating lanes that each order needs to enter; where lane i represents the lane where the items contained in order i are located; and the future period of time is adjusted according to the specific scenario.

[0018] In some embodiments, in step S2, the order retention algorithm designed according to the order prediction model includes:

[0019] S21. At the current time point, use the order prediction model to obtain the prediction scores of all orders in the order pool; where the prediction score refers to the data label in step S13;

[0020] S22. Sort the orders in the current order pool according to the prediction scores and initialize a pointer to 0;

[0021] S23. If the sum of the pointer and the batch capacity is greater than the number of orders in the order pool, go to step S26; otherwise, go to S24. The batch capacity is defined as the maximum number of orders / items that can be accommodated in a batch.

[0022] S24. Calculate the difference in the predicted scores of the current pointer order and the order at the current pointer plus the batch capacity. If the difference is less than the threshold, go to step S25; otherwise, go back to step S23. The threshold is an algorithm parameter and is adjusted according to the actual scenario.

[0023] S25. Release the orders from the current pointer to the current pointer plus the batch capacity to the batch pooling, reset the pointer to 0, and go back to step S23.

[0024] S26. If the number of batches that can be formed in the batch pooling is less than the number of idle pickers, select the orders with low predicted scores in the order pool and supplement them to the batch pooling in sequence until the number of batches that can be formed is not less than the number of idle pickers.

[0025] S27. End the order reservation algorithm.

[0026] In some embodiments, in step S3, batch forming, sorting, allocation, and routing of the orders in the batch pooling according to the number of idle pickers in the system include:

[0027] S31. Solve the static order batch forming problem with the goal of the shortest picking distance.

[0028] S32. Sort the batches obtained by the batch forming algorithm according to a certain rule. The sorting rules include but are not limited to: the longest / short operation time, the earliest / latest generation time.

[0029] S33. Allocate the sorted batches to the pickers according to a certain allocation rule.

[0030] S34. Select a suitable routing algorithm to guide the pickers to complete the batch picking operation. The routing algorithms include but are not limited to: the shortest path algorithm, the S-shaped routing, the return routing, the maximum gap rule.

[0031] In some embodiments, in step S31, the algorithms that can be used to solve the static order batch forming problem include: integer programming method, column generation algorithm, heuristic algorithm, intelligent algorithm.

[0032] In some embodiments, in step S33, the allocation rule is at least one of the following rules: preferentially allocate batches to the picker with the shortest total working time, preferentially allocate batches to the picker who first completed the previous task.

[0033] In some embodiments, the algorithm parameters in step S2 include: item demand distribution, maximum batch capacity, number of pickers, order arrival distribution, prediction time span, release threshold.

[0034] In some embodiments, the heuristic algorithm metrics in step S3 include: order turnover time, longest completion time, and total service time.

[0035] The present invention also provides an online order batching system based on an order prediction reservation mechanism, including a processor and a memory. A computer program is stored in the memory, and the computer program can be processed to execute the above method.

[0036] The present invention has the following beneficial effects:

[0037] Based on machine learning methods, the present invention designs an order reservation strategy based on machine learning and a corresponding dynamic order batching method, that is, uses machine learning to predict order characteristics within a certain period in the future. Before batching, orders with optimization potential are reserved in advance and not immediately released in batches, so as to achieve a certain degree of global optimization. Finally, through subsequent batching, sorting, allocation, and routing algorithms, the batching of orders in the batching pool, setting the order of batch issuance, allocating batches to pickers, and guiding the routing of pickers are completed, and the order picking operation is completed. The present invention improves the order turnover rate and personnel idle time while optimizing the picking cost, and improves work efficiency. The present invention is applicable to the actual operation of a warehousing system with high immediacy requirements and a person-to-goods warehousing system with online order arrivals, and reduces the warehouse picking labor cost and improves the service level compared with the prior art. Description of the Drawings

[0038] Figure 1 is a flowchart of the method for online order batching based on an order prediction reservation mechanism in an embodiment of the present invention;

[0039] Figure 2a is an order schematic diagram of the online order batching method without using an order prediction reservation strategy in an embodiment of the present invention;

[0040] Figure 2b is an order schematic diagram of the online order batching method based on an order prediction reservation strategy in an embodiment of the present invention;

[0041] Figure 3 is a comparison diagram of the feature importance of the machine learning model in an embodiment of the present invention;

[0042] Figure 4 is a schematic diagram of the method for online order batching based on an order prediction reservation mechanism in an embodiment of the present invention;

[0043] Figure 5 is a heat map of the sensitivity analysis of the release threshold and the number of pickers in an embodiment of the present invention. Detailed Embodiments

[0044] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0045] As Figure 1 shown, an online order batching method based on an order prediction reservation mechanism provided by an embodiment of the present invention includes the following steps:

[0046] S1. Based on a machine learning method, predict future orders according to roadway similarity, and use order data to train an order prediction scoring model.

[0047] The arriving orders in the current cycle are orders 1, 2, and 3, and the arriving orders in the next cycle are orders 4 and 5. First, in the case of no reservation mechanism, in the current cycle, batch 1 consists of orders 1, 2, and 3; in the next cycle, batch 2 consists of orders 4 and 5. The routing of the two batches is as Figure 2a shown, and a total of 6 roadway crossings are required. In the case of the reservation mechanism, through prediction, it can be found that order 1 is close to orders 4 and 5, so order 1 can be reserved. Therefore, in the current cycle, batch 1 consists of orders 2 and 3; in the next cycle, batch 2 consists of orders 1, 4, and 5. The routing is as Figure 2b shown, and a total of 4 roadway crossings are required. Comparing the two situations, applying the order reservation mechanism reduces the number of roadway crossings by 2 times, reducing the walking distance and order turnover time. The scenario considered in the embodiment of the present invention is the picking requirement of a one-person arrival warehouse, which adopts the method of batch picking, that is, multiple orders are packed into one batch and converted into a picking list for the picker to execute. The embodiment of the present invention is applicable to the situation where orders arrive online, that is, orders gradually arrive at the warehouse over time. By predicting future orders, a certain degree of global optimization can be achieved by reserving some potential orders. Order data can select the real data of the warehouse and be used to train the model after cleaning.

[0048] S11. Order data requirements. The training data of the model is the order information in the warehousing system, including: order arrival time, items in the order, item locations, and item quantities.

[0049] S12. According to the order data, construct the features of the data. The features include: order arrival time (continuous value); the roadway location of the goods included in the order (0-1 value), where 0 indicates that there are no items to be picked in the order in this roadway, and 1 indicates that there are items to be picked in the order in this roadway; the quantity of items in the order.

[0050] S13. Construct data tags based on the order data. The tag is defined as the sum of the similarities between the current order and the arriving orders in a future period. Further, the order similarity is defined as: |lane i ∩ lane j| / |lane i ∪ lane j|, that is, the number of lanes that both orders need to enter divided by the number of non-repeating lanes that each order needs to enter. Here, lane i represents the lane where the items in order i are located, and lane j represents the lane where the items in order j are located. Among them, the future period is adjusted according to the specific scenario. In this method, this tag is called the prediction score. Calculate the prediction score for all order data.

[0051] S14. Divide the data with constructed features and labeled tags into a training set and a test set. Select a suitable machine learning regression model, including but not limited to: tree models, ensemble models, deep learning models. Select a suitable model evaluation criterion, such as mean squared error. Train and validate the model on the training set and evaluate the model on the test set.

[0052] As Figure 3 shown, according to the training results of a certain warehouse, the arrival time and type A lanes are important features. The former reflects that the model can predict the time distribution of order arrivals, that is, the prediction scores of different orders at different times may be different; the latter reflects the model's preference for the importance of high-frequency lanes, because high-frequency lanes often appear in orders, so the model tends to give high scores to orders with high-frequency lanes. Through the trained order prediction scoring model, input the current order pool structure, and it is possible to predict the future arriving order pool structure in the current cycle. Retain the orders similar to the future orders until the future orders arrive and are grouped together, achieving a certain degree of global optimization.

[0053] S2. Score the orders in the order pool based on the order prediction model, design an order retention algorithm, and decide whether to retain the order in the order pool or release it to the batch pool. In order to minimize costs while ensuring the order turnover rate, the retained orders should meet the following conditions: 1. Can form a better batch with future orders, that is, the model gives a higher prediction score; 2. The number of similar orders in the current order pool is less than the batch capacity. If the number of similar orders has reached a batch capacity, they can be released together; 3. Try not to let the picker be idle. After releasing the orders that meet the batch capacity, when the picker is idle, release the orders with little potential (i.e., low prediction scores). Based on the above logic, the designed order retention algorithm process is as follows:

[0054] S21. At the current time point, use the order prediction model to obtain the prediction scores of all orders in the order pool, where the prediction score refers to the data tag in step S13;

[0055] S22. Sort the orders in the current order pool according to the predicted scores, and initialize a pointer to 0;

[0056] S23. If the sum of the pointer and the batch capacity is greater than the number of orders in the order pool, go to S26; otherwise, go to S24. The batch capacity is defined as the maximum number of orders / items that can be accommodated in a batch.

[0057] S24. Calculate the difference in predicted scores between the order at the current pointer and the order at the current pointer plus the batch capacity. If the difference is less than the threshold, go to S25; otherwise, go back to S23. The threshold is an algorithm parameter and can be adjusted according to the actual scenario.

[0058] S25. Release the orders from the current pointer to the current pointer plus the batch capacity to the batch pooling, reset the pointer to 0, and go back to S23;

[0059] S26. If the number of batches that can be formed in the batch pooling is less than the number of idle pickers, select the orders with low predicted scores in the order pool and supplement them to the batch pooling in turn until the number of batches that can be formed is not less than the number of idle pickers;

[0060] S27. End the order retention algorithm.

[0061] As Figure 4 shown, the order retention algorithm starts when new orders arrive. First, the order scoring model scores the orders in the order pool, and then the order pool is divided into two parts according to the above order retention algorithm, with one part retained in the current order pool and the other part released to the batch pooling. By controlling the threshold parameter, the mandatory nature of order retention can be achieved, and for different systems, it can be optimized by adjusting the threshold parameter.

[0062] S3. According to the number of idle pickers in the system, batch, sort, allocate, and route the orders in the batch pooling, including:

[0063] S31. With the goal of the shortest picking distance, solve the static order batching problem. The algorithms that can be used include, but are not limited to: integer programming method, column generation algorithm, heuristic algorithm, intelligent algorithm, etc. Specifically, a suitable algorithm can be selected according to the timeliness requirements;

[0064] S32. Sort the batches obtained by the batching algorithm according to a certain rule and send them to the pickers. The sorting rules include, but are not limited to: longest / short operation time, earliest / latest generation time;

[0065] S33. Allocate the sorted batches to the pickers according to a certain rule. The allocation rules include, but are not limited to: preferentially allocate batches to the picker with the shortest total working time, preferentially allocate to the picker who completed the previous task first;

[0066] S34. Select a suitable routing algorithm to guide the picker to complete the batch picking operation. The routing algorithms include but are not limited to: the shortest path algorithm, S-shaped routing, return routing, and maximum gap rule.

[0067] Orders in the batch pooling will be converted into batches through the order batching algorithm with the goal of minimizing the picking cost. Subsequently, the generated batches will be sorted by the batch sorting algorithm. Next, the sorted batches will be distributed to the pickers through the batch allocation algorithm. Finally, through the routing algorithm, the picker enters the aisle to collect all the orders in the batch and completes the picking operation.

[0068] The time points defined in the embodiments of the present invention are as follows: Order arrival time: the arrival time when the order arrives at the warehouse; Order / batch start time: the time when the batch where the order is located is assigned to the picker and the picker starts picking; Order completion time: the time when the batch where the order is located is collected by the picker and returns to the workbench.

[0069] The time periods defined in the embodiments of the present invention are as follows: Order waiting time: the waiting time of the order in the system before starting service, that is, the order start time minus the order arrival time; Order / batch service time: the time consumed for the batch where the order is located to be executed and completed by the picker, including walking time and picking time; Order turnaround time: the passing time of the order in the system, that is, the order completion time minus the order arrival time.

[0070] The optimization metrics considered in the embodiments of the present invention are the average order turnaround time and the total batch service time, which represent the service level and picking cost respectively.

[0071] As Figure 5 shown, for the order data of a certain warehouse, a sensitivity analysis is performed on the number of pickers and the order release threshold. The heat map reflects the relationship between the pickers and the order release threshold: when the number of pickers is small, choosing a smaller release threshold algorithm performs better; when the number of pickers is large, choosing a larger release threshold algorithm performs better. There is a positive correlation between the two.

[0072] In the embodiments of the present invention, a machine learning method is first used to implement the prediction of orders, and the order arrival time and the lane where the item is located are used as features. The lane similarity is used to characterize the prediction index to achieve the purpose of optimizing the subsequent picking cost. Based on the order prediction model, orders in the order pool are scored, and an order retention algorithm is designed. A group of orders that reach the upper limit of the batch capacity and the difference in predicted scores is less than the set threshold is released into the batch pool, and then the orders with low scores are released to idle pickers, so that the embodiments of the present invention optimize the picking cost while improving the order turnover rate and the idle time of personnel, and improve work efficiency. Finally, through subsequent batching, sorting, allocation, and routing algorithms, the batching of orders in the batch pool, setting the order of batch issuance, allocating batches to pickers, and guiding the routing of pickers are completed, and the order picking operation is completed, so as to be applicable to the actual operation of a warehousing system with high real-time requirements, thereby reducing the warehouse labor cost and improving the order service level.

[0073] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0075] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to generate a computer-implemented process, thereby providing instructions for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps for the functions specified in one block or a plurality of blocks.

[0077] The above is a further detailed description of the present invention in combination with specific / preferred embodiments. It should not be considered that the specific implementation of the present invention is limited only to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several alternatives or modifications can be made to the described embodiments, and these alternative or modified forms should be regarded as falling within the protection scope of the present invention. In the description of this specification, the description with reference to the terms "embodiment", "some embodiments", "preferred embodiment", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. Without conflict, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions, and alterations can be made herein without departing from the scope of protection of the patent application.

Claims

1. Online order batching method based on order prediction reservation mechanism, characterized in that, it includes the following steps: S1. Based on machine learning methods, predict future orders according to roadway similarity, and use order data to train an order prediction scoring model; S2. Score the orders in the order pool based on the order prediction scoring model, design an order reservation algorithm, and decide whether to retain the order in the order pool or release it to the batching pool; S3. According to the number of idle pickers in the system, use heuristic algorithms to batch, sort, allocate, and route the orders in the batching pool; Step S1 includes: S11. Provide order data, that is, the training data of the model, which is the order information in the warehousing system, including: order arrival time, items in the order, item location, and item quantity; S12. According to the order data, construct the features of the data; the features include: order arrival time, which is a continuous value; the roadway location of the goods contained in the order, which is a 0-1 value, 0 indicates that there are no items to be picked in this roadway for the order, and 1 indicates that there are items to be picked in this roadway for the order; the quantity of items in the order; S13. According to the order data, construct the data label; the label is defined as: the sum of the similarities between the current order and the orders arriving in the future for a period of time; S14. Divide the data with constructed features and labeled data into a training set and a test set, select appropriate machine learning regression models, including: tree models, ensemble models, deep learning models; select appropriate model evaluation criteria: mean square error; train and validate the model on the training set, and evaluate the model on the test set; In the said step S2, the order reservation algorithm designed according to the order prediction scoring model includes: S21. At the current time point, use the order prediction scoring model to obtain the prediction scores of all orders in the order pool; where the prediction score refers to the data label in step S13; S22. Sort the orders in the current order pool according to the prediction scores, and initialize a pointer to 0; S23. If the sum of the pointer plus the batch capacity is greater than the number of orders in the order pool, go to step S26, otherwise go to S24, and the batch capacity is defined as the maximum number of orders / items that can be accommodated in a batch; S24. Calculate the difference in the prediction scores of the order at the current pointer and the order at the current pointer plus the batch capacity. If the difference is less than the threshold, go to step S25, otherwise go back to step S23. The threshold is an algorithm parameter and is adjusted according to the actual scenario; S25. Release the orders from the current pointer to the current pointer plus the batch capacity to the batching pool, reset the pointer to 0, and go back to step S23; S26. If the number of batches that can be formed in the batching pool is less than the number of idle pickers, select the orders with low prediction scores in the order pool and supplement them to the batching pool in turn until the number of batches that can be formed is not less than the number of idle pickers; S27. End the order reservation algorithm.

2. The online order batching method based on order prediction reservation mechanism according to claim 1, characterized in that, The definition of the order similarity is: |lane i ∩ lane j| / |lane i ∪ lane j|, that is, the number of lanes that two orders need to enter divided by the number of non-repeating lanes that the two orders need to enter respectively; where lane i represents the lane where the items included in order i are located; where the future period is adjusted according to the specific scenario.

3. The online order batching method based on the order prediction reservation mechanism as described in claim 1, characterized in that, in step S3, according to the number of idle pickers in the system, batching, sorting, allocating, and routing the orders in the batching pool includes: S31. Taking the shortest picking distance as the goal, solving the static order batching problem; S32. Sorting the batches obtained by the batching algorithm according to a certain rule, and the sorting rules include: the longest / shortest operation time, the earliest / latest generation time; S33. Allocating the sorted batches to pickers according to a certain allocation rule; S34. Selecting a suitable routing algorithm to guide the pickers to complete the batch picking operation, and the routing algorithms include: the shortest path algorithm, the S-shaped routing, the return routing, and the maximum gap rule.

4. The online order batching method based on the order prediction reservation mechanism as described in claim 3, characterized in that, in step S31, the algorithms that can be used when solving the static order batching problem include: integer programming method, column generation algorithm, heuristic algorithm, intelligent algorithm.

5. The online order batching method based on the order prediction reservation mechanism as described in claim 3, characterized in that, in step S33, the allocation rule is at least one of the following rules: preferentially allocate batches to the picker with the shortest total working time, preferentially allocate batches to the picker who first completes the previous task.

6. The online order batching method based on the order prediction reservation mechanism as described in any one of claims 1-3, characterized in that, the algorithm parameters in step S2 include: item demand distribution, maximum batch capacity, number of pickers, order arrival distribution, prediction time span, release threshold.

7. The online order batching method based on the order prediction reservation mechanism as described in any one of claims 1-3, characterized in that, the heuristic algorithm metrics in step S3 include: order turnover time, longest completion time, total service time.

8. An online order batching system based on the order prediction reservation mechanism, including a processor and a memory, characterized in that, the memory stores a computer program, and the computer program can be processed to execute the method as described in any one of claims 1-7.

Images