KR102316149B1 - Systems and methods for optimizing delivery route - Google Patents

Abstract

The present invention predicts the optimal delivery route based on the delivery person, the restaurant location, the delivery location and the cooking time of the food, so that when the delivery person arrives at the restaurant, the cooking of the food is not completed, so that the waiting time can be saved. An optimization system and method, wherein the service target area is divided into a plurality of areas by dividing the service target area into a grid of the same size A path optimization system and method comprising a receiver, a cooking time estimator for estimating a cooking time of a menu, and an optimal path estimator for calculating an optimal path based on region-specific transit time information and cooking time information.

Description

{Systems and methods for optimizing delivery route}

The present invention relates to a system and method for optimizing a route, and more particularly, predicting an optimal delivery route based on the delivery person, the location of the restaurant, the location to be delivered, and the cooking time of the food. It relates to a system and method for optimizing a route that can save time waiting for not completed.

Currently, the delivery service is divided into self-delivery using the store's own delivery man and delivery service through a delivery agency. There is a high probability that self-delivery agents, delivery agents, or delivery will be delayed or unable to be performed during times when orders are rushed, and there is an idle manpower for delivery during times when there are no orders. In addition, if you request a delivery service, if the delivery person arrives at the restaurant after cooking is complete, the food may get cold. If you arrive at the restaurant before cooking is complete, you may waste time. An efficient management system is required.

Conventionally, Korea Patent No. 10-1371091 "food delivery agency system" selects the delivery person with the shortest delivery time by combining the distance between the restaurant and the delivery person and the distance between the delivery person and the delivery requester among delivery workers belonging to the delivery agent. We are disclosing a technology for a food delivery agency system that can deliver food to a delivery requester in time. There is no information on whether the food has been cooked or not, so when the delivery person arrives at the restaurant, the food cools down and the taste of the food deteriorates. Otherwise, there is a problem in that time is wasted upon arrival if the cooking is not completed.

Therefore, there is a need for a system and method capable of optimizing the delivery route of the delivery man in consideration of the food cooking time.

Korean Patent No. 10-1371091

The present invention calculates the optimal route of delivery based on the delivery person's location, pickup location, drop location, delivery menu information, and cooking time information of the delivery menu. The purpose is to help you save money.

An object of the present invention is to more accurately predict the time it takes for a delivery person to deliver by predicting the transit time for each area based on the current location information and previous location information of a plurality of delivery men.

According to the present invention, when the current location information and the area corresponding to the previous location information are not adjacent areas, the number of areas included on a straight line from the previous location information to the current location information is divided by the time difference between the reception times to determine the transit time for each area. The purpose of this is to predict the transit time for the entire area to be delivered by the delivery person even if there is no information that the delivery person has passed through each area before.

An object of the present invention is to predict the transit time for each area by synthesizing the transit times of a plurality of delivery men passing through one area, so that the time for the delivery man to pass through each area can be more accurately predicted based on a plurality of information. .

The present invention is by excluding the transit time of the first delivery man when predicting the transit time of a specific area when the transit time for a specific area of the first delivery man among a plurality of delivery men is within a predetermined minimum time or exceeds a predetermined maximum time With the exception of inaccurate information, the purpose is to increase the accuracy when estimating the time passing through each domain.

The purpose of the present invention is to predict the transit time information for each area by inquiring the transit time information corresponding to the current situation information in the transit information database, so that the delivery man can more accurately predict the time the delivery man passes through each area according to the specificity of the situation. do it with

It is an object of the present invention to more accurately predict the time to pass through each area according to the specificity of the situation by including at least one of information indicating time, day of the week, weather, and holiday status.

The present invention confirms the area corresponding to the current location, the pickup location, and the drop location of the delivery man, and the optimal route from the area corresponding to the current location to the area corresponding to the pickup location and the drop location in the area corresponding to the pickup location An object of the present invention is to enable an optimal path to be calculated for each section by calculating an optimal path to a corresponding region.

The present invention confirms a plurality of delivery call information to check the current location of the delivery person, the area corresponding to each of the pickup location and drop location of a plurality of delivery calls, and calculates the overall optimal route, so that one delivery person performs a plurality of delivery orders aimed at making it possible.

The present invention predicts the cooking time based on the store information corresponding to the menu information included in the delivery call information, the material information, and the current kitchen state information, so that the delivery person can minimize the waiting time for the cooking time in the store. do it with

In order to achieve this object, the route optimization system according to an embodiment of the present invention divides the service target area into a grid of the same size and divides the service target area into a plurality of areas, predicting the passage time of each divided area. A delivery call receiving unit for receiving delivery call information including a transit time prediction unit, a pickup location, a drop location and delivery menu information, a cooking time prediction unit for predicting a cooking time of a menu corresponding to the delivery menu information, and the received delivery The call may include an optimal path calculating unit that calculates an optimal path based on the transit time information for each region predicted by the transit time prediction unit and the cooking time information predicted by the cooking time prediction unit.

In addition, further comprising a location information receiver for receiving the location information of a plurality of delivery men, the transit time prediction unit current location information of one of the plurality of delivery men received in the location information receiver and previous location information received in the previous period It may be configured to predict the passage time for each area based on the .

In addition, when the area corresponding to the current location information and the previous location information is not an adjacent area, the transit time prediction unit calculates the number of areas included on a straight line from the previous location information to the current location information. It may be configured by dividing the time difference between a time point at which information is received and a time point at which the current location information is received to predict the transit time for each area.

In addition, the transit time prediction unit may be configured to predict the transit time for each area by synthesizing the transit times of a plurality of delivery men passing through the one area.

In addition, the transit time prediction unit predicts the transit time of the specific area when the transit time for a specific area of the first delivery man, which is any one of the plurality of delivery men, is within a predetermined minimum time or exceeds a predetermined maximum time. When it may be configured to exclude the passage time of the first delivery man.

In addition, the transit time prediction unit stores the transit time information for each area estimated based on the received current location information of one of the plurality of delivery men and the previous location information received in the previous period in correspondence with the situation information to the transit information database. and predicting transit time information for each area by inquiring transit time information corresponding to current situation information in the pass information database.

In addition, the context information may be configured to include at least one or more of information indicating time, day of the week, weather, and whether or not there is a public holiday.

In addition, the optimal path calculating unit checks the received delivery call information,

Check the area corresponding to the current location of the delivery man, the pickup location, and the drop location, respectively, and the optimal route from the area corresponding to the current location to the area corresponding to the pickup location and the area corresponding to the pickup location. It may be configured by calculating the optimal path to the area corresponding to the drop position to calculate the overall optimal path.

In addition, the optimal path calculating unit confirms the plurality of delivery call information received from the delivery call receiving unit, and confirms the area corresponding to the current location of the delivery person, the pickup location and the drop location of the plurality of delivery calls, respectively, to determine the overall optimal route It may be configured to calculate.

In addition, the cooking time prediction unit may be configured to predict the cooking time based on store information corresponding to the menu information included in the delivery call information, material information, and current kitchen state information.

The present invention calculates the optimal route of delivery based on the delivery person's location, pickup location, drop location, delivery menu information, and cooking time information of the delivery menu. to save money.

The present invention makes it possible to more accurately predict the time it takes for the delivery person to deliver by predicting the transit time for each area based on the current location information and previous location information of a plurality of delivery men.

According to the present invention, when the current location information and the area corresponding to the previous location information are not adjacent areas, the number of areas included on a straight line from the previous location information to the current location information is divided by the time difference between the reception times to determine the transit time for each area. By predicting , it is possible to predict the transit time for the entire area to be delivered by the delivery person even if there is no information that the delivery person has passed through each area before.

The present invention synthesizes the transit time of a plurality of delivery men passing through one area and predicts the passage time for each area, so that the time for the delivery man to pass through each area can be more accurately predicted based on a plurality of information.

The present invention is by excluding the transit time of the first delivery man when predicting the transit time of a specific area when the transit time for a specific area of the first delivery man among a plurality of delivery men is within a predetermined minimum time or exceeds a predetermined maximum time With the exception of inaccurate information, it is possible to increase accuracy when estimating the time to pass through each domain.

The present invention predicts the transit time information for each area by inquiring the transit time information corresponding to the current situation information in the transit information database, so that the delivery man can more accurately predict the time it takes to pass through each area according to the specificity of the situation.

According to the present invention, the situation information includes at least one of information indicating time, day of the week, weather, and whether it is a public holiday, so that the time passing through each area can be more accurately predicted according to the specificity of the situation.

The present invention confirms the area corresponding to the current location, the pickup location, and the drop location of the delivery man, and the optimal route from the area corresponding to the current location to the area corresponding to the pickup location and the drop location in the area corresponding to the pickup location By calculating the optimal path to the corresponding area, the optimal path can be calculated for each section.

The present invention confirms a plurality of delivery call information to check the current location of the delivery person, the area corresponding to each of the pickup location and drop location of a plurality of delivery calls, and calculates the overall optimal route, so that one delivery person performs a plurality of delivery orders make it possible

The present invention predicts the cooking time based on the store information corresponding to the menu information included in the delivery call information, the material information, and the current kitchen state information, so that the delivery person can minimize the waiting time for the cooking time in the store.

1 is a diagram illustrating an example of delivery by a delivery person according to an optimized delivery route according to an embodiment of the present invention.
2 is a diagram showing the configuration of a path optimization system according to an embodiment of the present invention.
Figure 3 is a view showing an example of the time saved by predicting the cooking time according to an embodiment of the present invention, the delivery man arrives at the time when the cooking is finished.
4 is a diagram illustrating an example of dividing a service target area into a grid of the same size according to an embodiment of the present invention.
5 is a diagram illustrating an example of a graph for predicting a passage time of a region according to an embodiment of the present invention.
6 is a diagram illustrating an example of information predicting a passage time of a region according to an embodiment of the present invention.
7 is a diagram illustrating an example of calculating the entire route by calculating the optimal route from the pickup location to the drop location according to an embodiment of the present invention.
8 is a diagram illustrating an example of context information stored in a pass information database according to an embodiment of the present invention.
9 is a view showing an example of delivery by a delivery man according to an optimized delivery route in consideration of the preferred area of the delivery man according to an embodiment of the present invention.
10 is a flowchart illustrating a path optimization method according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, in describing the embodiments of the present invention, specific numerical values are merely examples and the scope of the invention is not limited thereby.

The route optimization system according to the present invention may be configured in the form of a server having a central processing unit (CPU) and a memory (Memory) and connectable to other terminals through a communication network such as the Internet. However, the present invention is not limited by the configuration of the central processing unit and the memory. In addition, the folder search system according to the present invention may be physically configured as one device or may be implemented in a distributed form among a plurality of devices, so the present invention is not limited by the configuration of such a physical device.

1 is a diagram illustrating an example of delivery by a delivery person according to an optimized delivery route according to an embodiment of the present invention.

Delivery man refers to a subject who picks up goods from a specific place and then drops them to a specific place. can represent Currently, delivery agents can be divided into delivery agents, delivery workers owned by the store, and individual delivery workers.

When a delivery service is requested from a delivery agent at the current pickup location using the delivery service, the order from the restaurant is sent to multiple delivery workers, and the first delivery person who accepts the order takes the order. It is proceeding in a combat dispatch method. In this case, the delivery person accepts the order by looking at only the pick-up and drop-off places, and it is a method of moving with one order that can usually perform one pick-up and one drop. Currently, most delivery agencies are using this method, and this causes several problems.

The first is the issue of safety. In order to receive an order and perform delivery in a competitive combat dispatch environment, you have to press the accept button before other riders, and you have to make an effort to press the button not only at a stop, but also while driving. To solve this problem, instead of pressing a button, it is possible to deliver dispatch information by voice and decide whether to accept it by voice, but there is a problem in that it is difficult to deliver accurate information and the rider has to optimize the route in his/her head while driving. Because it is a combat dispatch environment, there is a high probability of accepting it rather than an accurate judgment. Because of this, various problems may occur, such as missing an opportunity to optimize the route, and the food being delivered or the food to be delivered is cold or unreasonable operation occurs.

The second is the problem of picking. In combat dispatch, the location of the order store and delivery destination is exposed, and the deliveryman can decide whether to accept it according to his or her choice. To compensate for this, compensation is required according to the hourly wage operation or the distance traveled from the location of the delivery person to the store or delivery location.

The third is the problem of path optimization. Processing multiple orders at once, rather than processing one order at a time, can be more efficient as a rush of orders can be processed faster and another order on the path can be processed by one delivery person. Even delivery workers who are paid per case can maximize profits by processing multiple orders in a short time. However, as the number of destinations increases, the number of cases for calculating the optimal order of visits increases exponentially as the number of destinations increases. Therefore, it is not used well by current delivery agencies. In order to solve the optimization problem, the connection of data is important, and it is necessary to clearly know the destination, the time required between the visits, and the time spent in the visited place, and it is necessary to secure a business for collecting related data.

In addition, it is a fatal problem for the business to have the store's own deliveryman or delivery agent, or barriers against idle times when no orders are placed and peak times when orders are crowded There is a problem of idle and peak hours. And there is the delivery area problem, which is a problem of regional division for calculating the delivery fee.

As such, if a delivery person optimizes the route when delivering, there is an advantage that multiple deliveries can be made in the same time as one delivery.

For example, as shown in FIG. 1 , when a delivery order requested by drop 1 and drop 2 is received at pickup 1 and pickup 2, which are pickup locations, delivery person 1 is close to pickup 1 and delivery person 2 is close to pickup 2. However, if it is calculated that the cooking time at Pickup 1 ends and the cooking time at Pickup 2 is completed at the time of moving to Pickup 2 by calculating the optimized route, and it is calculated that the dish can be taken immediately, Delivery Man 1 will pick up Pickup 1 and Pickup 2 It can be efficient to pick up food at 2 and visit drop 1 and drop 2 sequentially. This has the advantage of saving time and manpower compared to delivery by delivery person 1 and delivery person 2 respectively, and when pickup 2 requests an order from delivery person 2 because pickup 2 is close to delivery person 2, delivery person 2 makes another delivery. It can be more inefficient. In this way, it has the advantage of maximizing profits if one delivery person handles multiple delivery orders by calculating the optimized route using the order receipt time, delivery time to the kitchen, cooking start time, and cooking completion time. .

2 is a diagram illustrating a configuration diagram of a path optimization system 200 according to an embodiment of the present invention.

The route optimization system 200 according to the present invention includes a region division unit 201, a location information reception unit 202, a transit time prediction unit 203, a delivery call reception unit 204, a cooking time prediction unit 205, and an optimal path. It may be composed of the operation unit 206, and each component may be a software module that operates in the same computer system physically, and is configured such that two or more physically separated computer systems can operate in conjunction with each other. may be, and various embodiments including the same function belong to the scope of the present invention.

The area dividing unit 201 divides the service target area into a grid of the same size and divides the service target area into a plurality of areas.

The service target area may be a location of a delivery person, a location of a restaurant, and an area where a delivery recipient is located. If the coordinates are approximated by reducing the number of decimal places representing latitude and longitude in the service target area, a shape close to a rectangle may appear, and one rectangular mirror can be divided into one area. If the coordinates are expressed to three decimal places, about 10 million squares can be created in Korea, and one square can be divided into one area, and one area can be about 90 meters wide and 110 meters long. The size of the region can be divided based on the number of decimal places of the coordinates.

The location information receiving unit 202 receives the location information of a plurality of delivery men.

The location information receiving unit 202 may receive the current location information of the delivery man on the move or the waiting delivery man.

The transit time prediction unit 203 predicts the transit time of each divided region.

The transit time prediction unit 203 may predict the time it takes for the delivery person to pass through one area divided by the area division unit 201 . The transit time for the delivery man to pass through each area may be different, and the time for the delivery man to pass through each area can be predicted. Currently, the traffic information provided by the T-map or navigation system provides road traffic conditions based on automobiles. In the case of delivery, the majority of people travel by motorcycle, and the time required to pass through a specific area is usually reduced with the T-map or navigation provided. may be different. Therefore, a technology for predicting the time required for a delivery person to pass through a specific area using a motorcycle is required.

The transit time prediction unit 203 predicts the transit time for each area based on the current position information of one of the plurality of delivery men received from the position information receiving unit 202 and the previous position information received in the previous period.

It is possible to predict the transit time for each area based on the time when the delivery man passes each area, and the location when the delivery man passes one area in the present or in the past is received and the time passed through that area is received. It is possible to predict the transit time for each area. In addition, if a delivery person is currently passing a specific area, it is possible to predict the transit time of the area based on the time information that the delivery person passed through the area in the past and the time information when another delivery person passed the area. If a large number of deliverymen are currently passing through each area, the current transit time for each area can be predicted, but if there are not many deliverymen currently passing each area, the transit time of each area can be predicted based on the records that have passed in the past. can

For example, if the current delivery person passes through a specific area 1 for delivery, the time it takes for the current delivery person to pass through area 1 is calculated based on the information that other delivery workers have passed through area 1 in the past based on the current location information. predictable.

When the area corresponding to the current location information and the previous location information is not an adjacent area, the transit time prediction unit 203 calculates the number of areas included on a straight line from the previous location information to the current location information after receiving the previous location information. By dividing the time difference between the time point and the time point at which the current location information is received, the transit time for each area is predicted.

If there is no information previously passed by another delivery person at the current location where the delivery person is trying to pass, if it is connected in a straight line with the point for the previous location closest to the current location where the delivery person is located, each area is based on the area included in the straight line. The transit time can be predicted. The position where the delivery person is currently located may be a start location, and the previous location adjacent to the current location of the delivery person may be an end location. When the start position and the end position are connected in a straight line, one or more regions may be included between the linearly connected regions, and the passage time for each region can be predicted based on the number of regions connected in a straight line.

There is a record of a delivery man moving from the start position to the end position in the past, but the area passed from the start position to the end position varies according to traffic conditions, so there may not be a predicted transit time for all areas that the delivery person is currently moving. Therefore, it is possible to estimate the area on a straight line from the start position to the end position based on the record that the delivery person has previously moved from the start position to the end position.

For example, if the time that the delivery person moves from the start position to the end position is 24 seconds regardless of the area the delivery person moves, the area included on the straight line may be 24 if displayed in a straight line from the start position to the end position. Since we can guess that each of the 24 areas can be passed in 24 seconds, we can predict that it takes 1 second to pass through each area. It is possible to more accurately predict the transit time for each area if the data of the delivery man who passed the start and end positions is accumulated.

The transit time prediction unit 203 predicts the transit time for each area by synthesizing the transit times of a plurality of delivery men passing through one area.

The transit time prediction unit 203 may predict the transit time for each area based on the accumulated transit time whenever a plurality of delivery personnel pass through one area. If the transit time of one area is predicted based on the data of multiple deliverymen rather than the data passed by one deliveryman, the transit time can be predicted more accurately because the transit time is predicted based on more data.

When the transit time prediction unit 203 predicts the transit time of a specific area when the transit time for a specific area of the first delivery person, which is any one of a plurality of delivery men, is within a predetermined minimum time or exceeds a predetermined maximum time The transit time of the first delivery man is excluded.

The transit time passing through one area may have an absolute average range of time that can physically act, and if it is within or exceeding the average range of time, the transit time prediction unit 203 calculates the time as one area. It can be an exception as data that is predicted to pass through. If the delivery person says that the transit time in a specific area is within the predetermined minimum time, it means that the delivery person passed the specific area too quickly. could have done In addition, if the delivery time in a specific area exceeds the predetermined maximum time, it means that the delivery person passed the specific area too late. This cannot be considered as transit time for the delivery movement and can therefore be recognized as an error. Data on transit times that are erroneous can be excluded from predicting transit times.

For example, if the average range of time suitable for passing through a specific area, area 2, is 20 seconds to 10 minutes for data from multiple delivery agents, delivery person 1 passes zone 2 in 10 seconds, and delivery person 2 It is said that it took 1 minute for Deliveryman 3 to pass through Zone 2, and 1 hour for Deliveryman 3 to go through Zone 2. The transit time prediction unit 203 may determine the time it takes for the delivery man 1 and the delivery man 3 to pass through the area 2 as an error or an exception and exclude it from predicting the transit time of the area 2, based on the transit time of the delivery man 2 The transit time of region 2 can be predicted.

The transit time prediction unit 203 stores the transit time information for each area estimated based on the current location information of one of a plurality of received delivery men and the previous location information received in the previous period in correspondence with the situation information to the transit information database. Then, the transit time information corresponding to the current situation information is inquired from the transit information database to predict the transit time information for each area.

The transit time prediction unit 203 may store the transit time information and context information for each area in correspondence with the passing time information for each area based on the current location information and the previous location information, and may predict the transit time information for each area based on this. have. The transit time prediction unit 203 may predict transit time information for a specific area based on the current location and previous location of the delivery man. When estimating the transit time, it can be stored along with information about the circumstances of the day.

For example, if deliveryman 1 passes through area 3 on holidays, holiday information and area 3 passing time can be stored together in the transit information database, and when delivery person 2 passes through area 3 on Monday, Monday information and area 3 passing time are stored together. It can be stored in the pass-through information database. In addition, the transit time of area 3 may be predicted based on the information that delivery person 1 and delivery person 2 passed through area 3 and stored in the passing information database. If deliveryman 3 passes through area 3 on the following Monday, the transit time required to pass through area 3 on Monday can be predicted from the transit information database.

The context information includes at least one of information indicating time, day of the week, weather, and whether it is a public holiday.

The time passing through the specific area may be traffic information, and the traffic information may be different depending on circumstances. Rainy days may take longer to pass through certain areas, and Monday mornings and Friday evenings may also take longer to pass. In the case of a busy downtown area on a public holiday, the transit time may take longer due to the large number of people, and in an area with a large number of companies, the transit time may take less time on a public holiday.

The delivery call receiving unit 204 receives delivery call information including pickup location, drop location, and delivery menu information.

The pick-up location may be a location of a place that provides delivery goods, and the drop location may be a location of a place where the delivery article is provided. It may be information about the delivery goods that the delivery person wants to deliver. The delivery call information may be information that a delivery person can move from a pickup location to a drop location with a delivery menu, and may be information requesting delivery to the delivery person.

The cooking time prediction unit 205 predicts the cooking time of the menu corresponding to the delivery menu information.

When the delivery item is food, the cooking time for the delivery menu corresponding to the delivery item may be different for each delivery menu. It may take 20 minutes for chicken delivery menu, 5 minutes for pork feet because it is prepared in advance, and 30 minutes for steamed chicken.

The cooking time prediction unit 205 predicts the cooking time based on the store information corresponding to the menu information included in the delivery call information, the material information, and the current kitchen state information.

The delivery call information includes a menu to be delivered by the delivery man, and the cooking time can be predicted based on store information, material information, and current kitchen state information corresponding to the menu information to be delivered by the delivery man. In the store information, depending on the cooking speed of the chef of the store, there may be a difference in the cooking time depending on the difference in cooking time for each store and whether there are a number of previous orders. Also, in the material information, there may be a difference in cooking time depending on whether the ingredients used for the menu have been purchased in advance at the store or whether the ingredients have been prepared in advance. Also, in the current kitchen state information, there may be a difference in cooking time depending on whether the kitchen tools can be used immediately or whether the heat for cooking is strong or weak, so that the cooking time can be predicted based on the information. In addition, the cooking time may include not only cooking time in the kitchen, but also cooking time in the kitchen and putting it in a packaging container for delivery by a delivery person.

For example, store A and store B are chicken stores, and store A starts with fried chicken first and preheated oil. Cooking can be finished, and when a user orders chicken with raw chicken stocked in the refrigerator, store B preheats the oil, cuts the raw chicken, makes a frying coat, coats the chicken with tempura and frying the chicken to cook the chicken in one word. This may take 30 minutes. As such, although store A and store B sell the same chicken, each store can predict the cooking time based on the current state.

The optimal path calculating unit 206 calculates an optimal path for the received delivery call based on the transit time information for each region predicted by the transit time prediction unit 203 and the cooking time information predicted by the cooking time prediction unit 205 . do.

With respect to the delivery call including the pickup location and the drop location information, the transit time prediction unit 203 may predict the transit time information of the pickup location and the drop location, and the delivery person receives the delivery menu from the pickup location at the current location and receives the drop location It is possible to calculate the optimal path to deliver the delivery menu to The optimal route calculating unit may calculate an optimal route corresponding to the delivery person in the most suitable position for delivery based on the transit time information and the cooking time information.

The optimal route calculating unit 206 checks the received delivery call information, checks the area corresponding to the current location, the pickup location, and the drop location of the delivery person, and extends from the area corresponding to the current location to the area corresponding to the pickup location. The optimal path is calculated by calculating the optimal path from the region corresponding to the optimal path and the pickup position to the region corresponding to the drop position.

The optimal route calculating unit 206 may calculate an optimized route from the current location to the pickup location by checking the area corresponding to the delivery person's current location, pickup location, and drop location to reach the drop location. The cooking time may be included when calculating the minimized route from the current location to the pick-up location and the drop location.

The optimal path calculating unit 206 may assign the most suitable delivery person in consideration of the current location, pickup location, and drop location with a plurality of delivery workers by checking the delivery call information.

For example, if deliveryman 1 is located at the current location where delivery person 1 is waiting, and it is necessary to get a delivery menu from pickup 1, which is the pickup location, and deliver it to drop 1, which is the drop location. An optimal route from pickup 1 can be created, and an optimal route from pickup 1 to drop 1 can be created. In this case, an optimal route including the time to cook the delivery menu in Pickup 1 can be created.

The optimal path calculating unit 206 checks the plurality of delivery call information received from the delivery call receiving unit 204, and confirms the area corresponding to the current location of the delivery person, the pickup location and the drop location of the plurality of delivery calls, and confirms the overall optimal Calculate the path.

The delivery call receiving unit 204 may receive a plurality of deliveries, and the optimal path calculating unit 206 may calculate an optimal path so that a single delivery person can deliver a plurality of delivery calls.

For example, if you receive two deliveries that are considered efficient to deliver two deliveries together, the first pickup location is Pickup1, the second pickup location is Pickup2, the first drop location is Drop1, and the second drop location is may be drop 2. It is possible to calculate an optimal route through which the delivery person can pass through all of pickup 1, pickup 2, drop 1 and drop 2 from the current location where the delivery person assigned to delivery is located. It takes 20 minutes from Pickup 1 to Drop 1, Pickup 2 is located on the way from Pickup 1 to Drop 1, and it is said that it takes about 1 minute to pass from Drop 1 to Drop 2. In addition, the distance from the current location of the delivery person to Pickup 1 takes about 5 minutes, the cooking time of the delivery menu at Pickup 1 takes 5 minutes, the cooking time of the delivery menu at Pickup 2 takes 15 minutes, and Pickup 2 at Pickup 1 takes 5 minutes. This may take up to 10 minutes. In this case, if delivery calls are received from Pickup 1 and Pickup 2 at the same time, cooking can be started at Pickup 1 and Pickup 2, and if the delivery person takes 5 minutes from the current location to Pickup 1 and cooks for 5 minutes at Pickup 1, The delivery man can receive the delivery menu from Pickup 1 immediately and depart for Pickup 2, and the delivery man arrives at Pickup 2 in 15 minutes, which is the total time the delivery person stops by Pickup 1 from the current location, receives the menu every month, and leaves for Drop 1. . After delivering the delivery menu of pickup 1 to drop 1, the delivery person may deliver the delivery menu of pickup 2 to drop 2. In this way, there is an advantage that time and cost can be saved when a single delivery person handles multiple delivery calls at one time.

Figure 3 is a view showing an example of the time saved by predicting the cooking time according to an embodiment of the present invention, the delivery man arrives at the time when the cooking is finished.

The optimal path calculating unit 206 calculates an optimal path for the received delivery call based on the transit time information for each region predicted by the transit time prediction unit 203 and the cooking time information predicted by the cooking time prediction unit 205 . do.

If the order delivery time is defined in 4 stages: 1. The delivery time from the current location of the delivery person to the pickup location, 2. The waiting time until departure after receiving the delivery menu from the pickup location, 3. The travel time from the pickup location to the drop-off location, 4 It is the waiting time until the next delivery call is received after delivery is completed at the drop location.

Looking at the conventional system 310 in FIG. 3, the location of the delivery man by time may be the time measured through GPS collection, and it can be divided into two types of movement and standby, and is expressed in blue and green, respectively, and expressed in blue. Travel time (movement time from ubiquitous location to pick-up location, travel time from pick-up location to drop-off location) is a time that can be reasonably estimated using simple statistical methods or services that provide traffic information by continuously collecting GPS data. The waiting time (waiting time from pick-up location to departure, waiting time from drop location to completion after delivery) expressed as , cannot be estimated simply using GPS data, and there is no separate service that provides time information. Most of the means of transportation used in the delivery industry are motorcycles, which have slightly different characteristics from automobiles, are less affected by traffic congestion, have ease of moving through narrow alleys, and are mainly automobiles because the delivery distance is within several kilometers. It may be burdensome or inappropriate to use a paid traffic information service for only the city.

The most of the delivery time is the waiting time from the pick-up location to the departure, and it may be the waiting time for the food to be cooked. If the delivery person picks up the food after it is already cooked, the food may get cold, and if you arrive at the store for pickup before cooking, it will affect the optimization of delivery. To reduce this time, it is necessary to accurately predict the cooking time, and linkage with a platform that can collect cooking data in stores is required.

For example, in the conventional system 310, the waiting time 312 from the pickup location to departure and the waiting time 314 from the drop location to completion after delivery in the four steps defining the order delivery time account for the largest portion. However, using the system of the present invention, since it provides an optimized route by calculating the cooking time, there is almost no waiting time 322 due to the cooking time at the pickup location, and it is suitable for delivery after completion of delivery at the drop location. It can be seen that the waiting time after delivery is significantly reduced because the delivery person is dispatched. Compared with the conventional system, the system of the present invention has the effect of reducing the cost because the time 325 is reduced by about three times.

4 is a diagram illustrating an example of dividing a service target area into a grid of the same size according to an embodiment of the present invention.

The area dividing unit 201 divides the service target area into a grid of the same size and divides the service target area into a plurality of areas.

The coordinates of latitude and longitude can be expressed with decimal points and signs, and although the earth is a sphere, in Korea, it is at a very small level compared to the surface area of the earth's sphere, so it can be safely assumed to be flat in consideration of some errors. Under this assumption, if the coordinates are approximated by reducing the number of decimal places representing latitude and longitude, a shape close to a rectangle can appear. One area 401 may have a width of about 90 meters and a length of about 110 meters, and may be suitable as a size of one area, but if it is necessary to manage it smaller, the number of digits after the decimal point may be increased.

For example, if the number of digits after the decimal point is one digit, the number of regions may be about 1000, and the size of the region may be 9 km horizontally and 11 km vertically. may be 0.9 km wide and 1.1 km long, and if the number of decimal places is three digits, the number of regions may be approximately 10 million, and the size of regions may be 90 m horizontal and 110 m vertical, and the number of decimal places is 4 digits. In this case, the number of regions may be about 1 billion, and the size of the regions may be 9 m in width and 11 m in height. In addition, when the number of digits after the decimal point is five digits, the number of regions may be approximately 100 billion, and the size of the region may be 90 cm in width and 110 cm in height.

5 is a diagram illustrating an example of a graph for predicting a passage time of a region according to an embodiment of the present invention.

When the area corresponding to the current location information and the previous location information is not an adjacent area, the transit time prediction unit 203 calculates the number of areas included on a straight line from the previous location information to the current location information after receiving the previous location information. By dividing the time difference between the time point and the time point at which the current location information is received, the transit time for each area is predicted.

It is difficult to predict the transit time of the area using the real-time location of the delivery man because an error may occur in receiving the delivery man's location information or the delivery man's communication cost may be excessively high. It can be difficult to update. In addition, if the delivery person's location information is continuously received, there is a possibility that the current location information and the previous location information are not adjacent, and there is no information about which area has passed between the start area and the end area.

For example, the point (0,0) is the start area 501, and the point (18,12) is the end area 502. If you look at a straight line passing from the start area 501 to the end area 502, on a plane A grid can be an area. At this time, all rectangles through which the straight line passes, that is, the area that should reflect the change in the value of the area. If the deliveryman's location information is received twice in total, once each in the start area 501 and the end area 502, the time difference at this time is 24 seconds. can The number of regions through which a straight line passes is 24, and the time passing through each region can be approximated to 1 second, so that the transit time of each region can be predicted.

6 is a diagram illustrating an example of information predicting a passage time of a region according to an embodiment of the present invention.

The transit time prediction unit 203 predicts the transit time of each divided region.

The transit time prediction unit 203 predicts the transit time for each area based on the current position information of one of the plurality of delivery men received from the position information receiving unit 202 and the previous position information received in the previous period.

When the area corresponding to the current location information and the previous location information is not an adjacent area, the transit time prediction unit 203 calculates the number of areas included on a straight line from the previous location information to the current location information after receiving the previous location information. By dividing the time difference between the time point and the time point at which the current location information is received, the transit time for each area is predicted.

The transit time prediction unit 203 predicts the transit time for each area by synthesizing the transit times of a plurality of delivery men passing through one area.

When the transit time prediction unit 203 predicts the transit time of a specific area when the transit time for a specific area of the first delivery person, which is any one of a plurality of delivery men, is within a predetermined minimum time or exceeds a predetermined maximum time The transit time of the first delivery man is excluded.

The transit time prediction unit 203 stores the transit time information for each area estimated based on the current location information of one of the plurality of received delivery men and the previous location information received in the previous period in correspondence with the situation information to the transit information database. Then, the transit time information corresponding to the current situation information is inquired from the transit information database to predict the transit time information for each area.

Each region may mainly include traffic information, and the transit time prediction unit 203 may predict the transit time of each region. The transit time prediction unit 203 may store in the transit information database together with situational information such as the time of the time, day of the week, weather and holidays based on the real-time current location of the delivery person, and the average transit time of each area as a color. can express

For example, it may be expressed that the closer to blue, the less time it takes to pass through the region, and the closer it is to red, the longer it takes to pass through the region. This may vary depending on the situation information. Areas that take a short transit time on weekdays may take longer transit times on public holidays This may change the transit time depending on the situation information.

7 is a diagram illustrating an example of calculating the entire route by calculating the optimal route from the pickup location to the drop location according to an embodiment of the present invention.

The transit time prediction unit 203 predicts the transit time of each divided region.

The optimal path calculating unit 206 calculates an optimal path for the received delivery call based on the transit time information for each region predicted by the transit time prediction unit 203 and the cooking time information predicted by the cooking time prediction unit 205 . do.

The optimal route calculating unit 206 checks the received delivery call information, checks the area corresponding to the current location, the pickup location, and the drop location of the delivery person, and extends from the area corresponding to the current location to the area corresponding to the pickup location. An optimal path is calculated by calculating the optimal path from the region corresponding to the optimal path and the pickup position to the region corresponding to the drop position.

The cost and route of moving from one area to another can be defined as 'area to area', and can be determined by adding up the values of all areas passed and recording the route passed. In addition, delivery men move relatively short distances of around 2-3 km, so most means are not cars, but motorcycles, which facilitate passage through alleys and are less affected by traffic. Therefore, the time error of the result of 'area to area' may not be large.

For example, when a plurality of adjacent areas are schematically viewed 710, each rectangle may be an area, and the start area 701 and the end area 702 may be the delivery person's current location, pickup location, and drop location. The number at the top of the region may be a time taken to pass through the region, and the number at the bottom may be the time taken to move from the start region 701 to the region. The lower end of the end area 702 can be obtained by calculating the number of the lower ends from the start area 701 to the end area 702 . In this case, the movable direction can be calculated by allowing the right, down, and right down directions, and can move in a direction in which the start area 701 and the end area 702 are closer. The optimal time taken to move from the start region 701 to the end region 702 may be calculated as 9 and an optimal path 703 may be calculated.

8 is a diagram illustrating an example of context information stored in a pass information database according to an embodiment of the present invention.

The transit time prediction unit 203 stores the transit time information for each area estimated based on the current location information of one of a plurality of received delivery men and the previous location information received in the previous period in correspondence with the situation information to the transit information database. Then, the transit time information corresponding to the current situation information is inquired from the transit information database to predict the transit time information for each area.

The context information includes at least one of information indicating time, day of the week, weather, and whether it is a public holiday.

The transit information database can store the estimated transit time information when the delivery person passes through one area and the situation information when the delivery person passes through one area in correspondence, and the transit time prediction unit allows the delivery person to pass through a specific situation or a specific area. In case of passing through, the transit time information can be predicted by inquiring the pass information database in which situation information and predicted transit time are stored.

The data delivered by the delivery man may be location information at that time, and when the transit time prediction unit 203 receives the location information, it may be stored in a Tonga information database that manages the real-time location of the delivery person, and based on the stored data, it is Transit time information can be predicted. In addition, it is possible to receive not only the location information of the time, but also the situation information of the time, and the situation information may be the day of the week, weather, and whether it is a public holiday.

The time information for a specific area may consist of the number of times it has passed the area and the time it has passed that area, and the transit time is stored as an average, and when a new value is received, a new average value is used using the existing number and time. can be calculated. The Zone DataEntry may be one piece of information about the area, and may be information on the means of transportation (motorcycle, automobile, etc.) in detail. In addition, information on each day of the week, holidays, and the last 30 minutes is included, and a time interval that can be stored separately for each day of the week can be set.

In addition, when information for a new area is entered, the date, time, whether there is a public holiday, and the last 30 minutes can all be updated. can be calculated with

9 is a view showing an example of delivery by a delivery man according to an optimized delivery route in consideration of the preferred area of the delivery man according to an embodiment of the present invention.

The optimal route calculating unit 206 is based on the received delivery call, the transit time information for each area predicted by the transit time prediction unit 203, the cooking time information predicted by the cooking time prediction unit 205, and the delivery person's preferred area. to calculate the optimal path.

The optimal path calculating unit 206 may receive the preferred area of the delivery man, and may receive the preferred area including the maximum area or the avoidance area to be delivered by the delivery man. A delivery man may have an area he/she wants to deliver to for any reason, and may only want to deliver within that area. Therefore, the region to which the delivery man wants to deliver can be set as the preferred region. The optimal route calculating unit 206 may reflect this and select a delivery person, and may calculate an optimal route as a route that does not deviate from the preferred area set by the delivery person. If the delivery person is out of the preferred area, another delivery call can be received and an optimal route can be calculated so that the delivery person can enter the delivery person's preferred area again.

For example, delivery person 1 has set a preferred area 901, and the optimal route calculating unit 206 selects an optimal path for delivery person 1 to route from pickup 4 to pickup 5, and drop 5 to drop 4, which is a drop location. can be calculated. However, drop 5 is out of the preferred area of delivery person 1 and delivery person 1 may be reluctant to deliver, but the optimal route calculating unit 206 calculates the optimal route of drop 6 in pickup 6, and the drop that is in the preferred area in pickup 6, not in the preferred area. It can be computed to return to 6. Even if delivery person 1 leaves the preferred area, an optimal path can be calculated so that delivery person 1 can enter the preferred area again.

10 is a flowchart illustrating a path optimization method according to an embodiment of the present invention.

The path optimization method according to the present invention is not limited only by the description below, and the technical description applied to the above-described path optimization system 200 may be applied and understood as it is.

In the delivery route optimization method operating in the route optimization system 200 having a central processing unit and a memory,

In the step of dividing the service target area into a plurality of areas by dividing the service target area into a grid of the same size (S1001), the area dividing unit 201 divides the service target area into a grid of the same size and divides the service target area into a plurality of areas.

In the step of receiving the location information, the location information receiving unit 202 receives the location information of a plurality of delivery men.

In the step of estimating the transit time ( S1002 ), the transit time of each region divided by the transit time prediction unit 203 is predicted.

Predicting the transit time (S1002) predicts the transit time for each area based on the current location information of one of the plurality of delivery men received from the location information receiving unit 202 and the previous location information received in the previous period.

In the step of estimating the transit time (S1002), when the area corresponding to the current location information and the previous location information is not an adjacent area, the number of areas included on a straight line from the previous location information to the current location information is calculated using the previous location information. The time difference between the received time and the current location information is divided by the time difference to predict the transit time for each area.

Predicting the transit time (S1002) predicts the transit time for each area by synthesizing the transit times of a plurality of delivery men passing through one area.

Predicting the transit time (S1002) is the passing time of the first delivery man, any one of the plurality of delivery men, for a specific area is within a predetermined minimum time or exceeds a predetermined maximum time, the passage time of the specific area The transit time of the first delivery person is excluded from the prediction.

Predicting the transit time (S1002) is a transit information database by matching the transit time information for each area estimated based on the current location information of one of the received plurality of deliverymen and the previous location information received in the previous cycle to the context information and predicts the transit time information for each area by inquiring the transit time information corresponding to the current situation information in the transit information database.

The context information includes at least one of information indicating time, day of the week, weather, and whether it is a public holiday.

In the step of receiving the delivery call information (S1003), the delivery call receiving unit 204 receives delivery call information including pickup location, drop location and delivery menu information.

In the step of estimating the cooking time ( S1004 ), the cooking time prediction unit 205 predicts the cooking time of the menu corresponding to the delivery menu information.

Predicting the cooking time (S1004) predicts the cooking time based on the store information corresponding to the menu information included in the delivery call information, the material information, and the current kitchen state information.

The step of calculating the optimal path (S1005) is a step of predicting the transit time information and cooking time for each area predicted in the step (S1002) of estimating the transit time for the delivery call received by the optimal path calculating unit 206 (S1004) ) calculates the optimal route based on the cooking time information predicted in

In the step of calculating the optimal route (S1005), the received delivery call information is checked, the area corresponding to the current location, the pickup location, and the drop location of the delivery person is checked, and the area corresponding to the pickup location in the area corresponding to the current location is checked. The entire optimal path is calculated by calculating the optimal path to the region and the optimal path from the region corresponding to the pickup position to the region corresponding to the drop position.

The step of calculating the optimal path (S1005) checks the plurality of delivery call information received from the delivery call receiving unit 204, and confirms the area corresponding to the current location of the delivery person, the pickup location and the drop location of the plurality of delivery calls, respectively. to calculate the overall optimal path.

Such a path optimization method may be implemented as an application or implemented in the form of program instructions that may be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination.

The program instructions recorded on the computer-readable recording medium are specially designed and configured for the present invention, and may be known and available to those skilled in the computer software field.

Examples of the computer-readable recording medium include hard disks, magnetic media such as floppy disks and magnetic tapes, optical recording media such as CDROM and DVD, and magneto-optical media such as floppy disks. , and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like.

Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

Although the above has been described with reference to the embodiments, those skilled in the art can variously modify and change the present invention within the scope without departing from the spirit and scope of the present invention described in the claims below.

201: area division unit 202: location information receiving unit
203: transit time prediction unit 204: delivery call receiving unit
205: cooking time prediction unit 206: optimal path calculation unit

Claims (21)

a location information receiving unit for receiving location information of a plurality of delivery men;
a region dividing unit dividing a service target region into a grid of the same size and dividing the service target region into a plurality of regions;
a transit time prediction unit for predicting a transit time of each of the divided areas based on the current position information of one of the plurality of delivery men received from the position information receiving unit and the previous position information received in a previous period;
a delivery call receiver for receiving delivery call information including pickup location, drop location and delivery menu information;
a cooking time prediction unit for predicting a cooking time of a menu corresponding to the delivery menu information based on store information, material information, and current kitchen state information corresponding to the delivery menu information included in the delivery call information; and
With respect to the received delivery call, an optimal path calculating unit that calculates an optimal path based on the transit time information for each region predicted by the transit time prediction unit and the cooking time information predicted by the cooking time prediction unit
including,
The transit time prediction unit
When the area corresponding to the current location information and the previous location information is not an adjacent area,
The number of areas included on a straight line from the previous location information to the current location information is divided by a time difference between a time point at which the previous location information is received and a time point at which the current location information is received to predict the transit time for each area,
The optimal path calculating unit
Receiving the information on each of the plurality of delivery workers, and when each delivery person leaves the preferred area, calculating an optimal route so that each delivery person can return to the preferred area based on the received preferred area information
A path optimization system characterized by delete delete delete delete According to claim 1,
The transit time prediction unit
The received current location information of one of the plurality of delivery men and the transit time information for each area estimated based on the previous location information received in the previous period are stored in the transit information database in correspondence with the situation information,
Predicting passage time information for each area by inquiring passage time information corresponding to the current situation information in the passage information database
A path optimization system characterized by 7. The method of claim 6,
The situation information includes at least one or more of information indicating time, day of the week, weather, and whether it is a public holiday
A path optimization system characterized by According to claim 1,
The optimal path calculating unit
By checking the received delivery call information,
Check the area corresponding to the current location of the delivery man, the pickup location and the drop location, respectively,
Calculating the optimal path from the area corresponding to the current location to the area corresponding to the pickup location and the optimal route from the area corresponding to the pickup location to the area corresponding to the drop location to calculate the overall optimal route
A path optimization system characterized by 9. The method of claim 8,
The optimal path calculating unit
By checking the plurality of delivery call information received in the delivery call receiving unit,
Calculating the overall optimal route by checking the area corresponding to the current location of the delivery person, the pickup location and the drop location of multiple delivery calls.
A path optimization system characterized by delete In the delivery route optimization method operating in a route optimization system having a central processing unit and a memory,
Receiving the location information of a plurality of delivery men in the location information receiving unit;
dividing the service target area into a plurality of areas by dividing the service target area into a grid of the same size;
Predicting the transit time of each of the divided areas based on the current location information of one of the plurality of delivery men received from the location information receiver in the transit time prediction unit and the previous location information received in the previous period;
Receiving delivery call information including pickup location, drop location and delivery menu information from the delivery call receiver;
Predicting the cooking time of the menu corresponding to the delivery menu information based on the store information, the material information, and the current kitchen state information corresponding to the delivery menu information included in the delivery call information in the cooking time prediction unit; and
Calculating an optimal route based on the transit time information for each region predicted by the transit time prediction unit and cooking time information predicted by the cooking time prediction unit with respect to the received delivery call by the optimum path calculating unit
including,
The step of estimating the transit time is
When the area corresponding to the current location information and the previous location information is not an adjacent area,
The number of areas included on a straight line from the previous location information to the current location information is divided by a time difference between a time point at which the previous location information is received and a time point at which the current location information is received to predict the transit time for each area,
The step of calculating the optimal path is
Receiving the information on each of the plurality of delivery personnel, and when each delivery person leaves the preferred area, calculating an optimal route so that each delivery person can return to the preferred area based on the received preferred area information
A path optimization method characterized by delete delete delete delete 12. The method of claim 11,
The step of estimating the transit time is
The received current location information of one of the plurality of delivery men and the transit time information for each area estimated based on the previous location information received in the previous period are stored in the transit information database in correspondence with the situation information,
Predicting passage time information for each area by inquiring passage time information corresponding to the current situation information in the passage information database
A path optimization method characterized by 17. The method of claim 16,
The situation information includes at least one or more of information indicating time, day of the week, weather, and whether it is a public holiday
A path optimization method characterized by 12. The method of claim 11,
The step of calculating the optimal path is
By checking the received delivery call information,
Check the area corresponding to the current location of the delivery man, the pickup location and the drop location, respectively,
Calculating the optimal path from the area corresponding to the current location to the area corresponding to the pickup location and the optimal route from the area corresponding to the pickup location to the area corresponding to the drop location to calculate the overall optimal route
A path optimization method characterized by 19. The method of claim 18,
The step of calculating the optimal path is
By checking the plurality of delivery call information received in the step of receiving the delivery call information,
Calculating the overall optimal route by checking the area corresponding to the current location of the delivery person, the pickup location and the drop location of multiple delivery calls.
A path optimization method characterized by delete A recording medium recording a program for executing the method of any one of claims 11 and 16 to 19.

Images