Description:TECHNICAL FIELD
The present disclosure relates generally to the field of data analysis and optimization and more specifically, to a system and method for optimizing delivery network using saturation or active state of source entities.
BACKGROUND
In today's interconnected world, the importance of ecommerce and logistics industries and associated technologies have significantly increased due to efficient and fast transfer of goods and services between merchants and end-users. Such transfer is achieved through delivery networks established by various ecommerce platforms. The delivery networks in ecommerce industries involve a chain of merchants, which are involved in manufacturing of goods, a plurality of consumers, which demand such goods and a chain of delivery personnel, which are responsible for transporting the goods from merchants to the consumers. In conventional delivery networks, orders for the goods are placed by the consumers on various online platforms and the corresponding platforms assign the orders to the delivery personnel for transporting the goods from the merchants to the consumers.
However, the problem associated with the conventional delivery networks is use of static allocation strategies that fails to adapt to changing operating conditions, such as change in capacity or availability of the merchants, change in types of orders, interaction with different types of entities (like warehouse, restaurants, supplier types), frequency of orders and the like, which results in inefficiencies and delays in the delivery process. Further, the conventional delivery networks involve allocation of the orders based only on geographical location of the delivery personnel. Therefore, there exist a technical problem of how to reduce delay in the delivery process by considering rapidly changing operating conditions and provide balanced distribution of workload among the delivery personnel. Further, conventional systems are limited to use of geocoding and geofencing, which uses GPS to create virtual boundaries around specific locations so that when a rider enters or leaves a geofenced area, a fleet manging entity can be notified.
Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks associated with the conventional viscosity density sensors.
SUMMARY
The present disclosure provides a system and method for optimizing delivery network using saturation or active state of source entities. The present disclosure provides a solution to the technical problem of how to reduce delay in the delivery process by considering rapidly changing operating conditions and provide balanced distribution of workload among the delivery personnel. An aim of the present disclosure is to provide a solution that overcomes at least partially the problems encountered in the prior art and provide an improved system and method for optimizing delivery network, which considers saturation or active state of source entities to allocate the workload to the delivery personnel.
One or more objectives of the present disclosure is achieved by the solutions provided in the enclosed independent claims. Advantageous implementations of the present disclosure are further defined in the dependent claims.
In one aspect, the present disclosure provides a system for optimizing delivery network. The system comprises a delivery network optimising (DNO) server comprising a hardware processor configured to demark a plurality of different geographical areas having a plurality of source entities and identify a plurality of incoming work instructions associated with the plurality of source entities from a plurality of end-user devices based on the demarked plurality of different geographical areas. Furthermore, the DNO server is configured to determine a frequency of the plurality of incoming work instructions for each of the plurality of source entities. In addition, the DNO server is configured to determine whether one or more source entities from the plurality of source entities are in a saturation state or an active state in an upcoming time period based on a trend in the frequency of the plurality of incoming work instructions received from the plurality of end-user devices and the determined one or more service domains. Further, the system comprises an allocation component, which is configured to automatically allocate the plurality of incoming work instructions to a plurality of riders based on the saturation state or active state of the plurality of source entities, the frequency of the plurality of incoming work instructions and the determined one or more service domains.
The system determines the saturation state or active state of the source entities in demarked geographical areas in advance and allocates the riders to serve the demarked geographical areas in proportion to the number of active and saturated source entities (more number of riders allocated to active areas than saturated areas), thereby provide balanced distribution of workload among the plurality of riders and reducing delay in the delivery process. The system further determines possibility of the saturation state and active state of the source entities with respect to type of work instruction (i.e., service domains), thereby enables the source entities to focus on a particular domain (having more demand in future) to maximize profits. The system changes the plurality of work instructions allocated to the plurality of riders in accordance with change in saturation and active states of the source entities, which enables the system to provide timely fulfilment of the work instructions from the source entities having dynamic fluctuation in operational state. The determination of the upcoming saturation or active state of source entities based on trends by the system allows for proactive decision-making for workload allocation. By predicting the workload and availability of source entities in advance, the system is configured to make informed decisions to balance the workload across the delivery network, preventing bottlenecks and ensuring timely deliveries.
Additionally, the system of the present invention achieves following technical effects and advantage over conventional systems.
a) The system's ability to demarcate different geographical areas within the delivery network allows for better organization and management of incoming work instructions. By defining specific regions, the system can streamline the allocation process and ensure that tasks are assigned to riders within their designated service domains.
b) Based on the trend in the frequency of incoming work instructions and the determined service domains, the system can proactively assess whether source entities are going to be in a saturation state or an active state in the upcoming time period. This assessment allows the system to understand the workload capacity of each source entity and make allocation decisions beforehand accordingly. By considering the saturation and activity levels, the system can avoid overburdening source entities or underutilizing available resources.
c) The system's taking into account geographical area demarcation, work instruction frequency analysis, saturation state and active state determination, and automated task allocation leads to improved allocation efficiency, optimal resource utilization, and adaptability in the delivery network optimization process.
In another aspect, the present disclosure provides a method for optimizing delivery network. The method comprises, in a delivery network optimising (DNO) server, demarking a plurality of different geographical areas having a plurality of source entities and identifying a plurality of incoming work instructions associated with the plurality of source entities from a plurality of end-user devices based on the demarked plurality of different geographical areas. Furthermore, the method comprises determining a frequency of the plurality of incoming work instructions for each of the plurality of source entities and determining one or more service domains of each work instruction from the plurality of incoming work instructions. In addition, the method comprises determining whether one or more source entities from the plurality of source entities are in a saturation state or an active state in an upcoming time period based on a trend in the frequency of the plurality of incoming work instructions received from the plurality of end-user devices and the determined one or more service domains. Further, the method comprises automatically allocating the plurality of incoming work instructions to a plurality of riders based on the saturation state or active state of the plurality of source entities, the frequency of the plurality of incoming work instructions and the determined one or more service domains.
The method achieves all the advantages and technical effects of the system of the present disclosure.
It is to be appreciated that all the aforementioned implementation forms can be combined. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. It will be appreciated that features of the present disclosure are susceptible to being combined in various combinations without departing from the scope of the present disclosure as defined by the appended claims.
Additional aspects, advantages, features, and objects of the present disclosure would be made apparent from the drawings and the detailed description of the illustrative implementations construed in conjunction with the appended claims that follow.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a network diagram of a system for optimizing delivery network, in accordance with an embodiment of the present disclosure;
FIG. 2 is a block diagram of a delivery network optimizing (DNO) server in a system for optimizing a delivery network, in accordance with another embodiment of the present disclosure;
FIG. 3 is a block diagram of a system for optimizing a delivery network, in accordance with another embodiment of the present disclosure;
FIG. 4 is a diagram of an exemplary scenario of the allocation of work instructions to a plurality of riders based on saturation state or active state of the source entities, in accordance with an alternative embodiment of the present disclosure; and
FIG. 5 is a flowchart of a method for optimizing a delivery network, in accordance with an embodiment of the present disclosure.
The summary above, as well as the following detailed description of illustrative embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the present disclosure, exemplary constructions of the disclosure are shown in the drawings. However, the present disclosure is not limited to specific methods and instrumentalities disclosed herein. Moreover, those skilled in the art will understand that the drawings are not to scale. Wherever possible, like elements have been indicated by identical numbers.
Embodiments of the present disclosure will now be described, by way of example only, with reference to the following diagrams wherein:
In the accompanying drawings, an underlined number is employed to represent an item over which the underlined number is positioned or an item to which the underlined number is adjacent. A non-underlined number relates to an item identified by a line linking the non-underlined number to the item. When a number is non-underlined and accompanied by an associated arrow, the non-underlined number is used to identify a general item at which the arrow is pointing.
DETAILED DESCRIPTION OF EMBODIMENTS
The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.
FIG. 1 is a network diagram of a system for optimizing a delivery network, in accordance with an embodiment of the present disclosure. With reference to FIG. 1, there is shown a system 100 for optimizing a delivery network 102. The system 100 includes a delivery network optimising (DNO) server 104, a plurality of end user devices 114, such as a first end-user 114A, a second end-user 114B up to Nth end-user 114N and the delivery network 102, which are communicatively coupled to each other through a communication network 112. The DNO server 104 further includes a hardware processor 106, a primary storage 106A and an allocation component 122. The hardware processor 106 is communicatively coupled to the primary storage 106A and the allocation component 122.
The present disclosure provides the system 100, which is adapted to optimize the delivery network 102 using saturation or active state of source entities, where the hardware processor 106 of the system 100 allocates work instructions to riders operating delivery vehicles based on the number of saturated or active source entities in different geographical areas. A source entity refers to type of client such as a corporate entity, an individual, a restaurant, a shop, a warehouse, a distribution center, a pharmacy, or an ecommerce platform such as Zomato, Swiggy, Blinkit or the like, which manufactures or supplies goods and/or services to consumers. An end-user device refers to an electronic device operated by an end-user, such as an individual, a group of individuals or a corporate entity. Each end-user transmits a work instruction through corresponding end-user device to one or more source entities. The work instruction refers to an order or requirement that is indicative of a demand for goods and/or services. The saturation state refers to a state of the source entity in which the corresponding source entity unable to receive or fulfil work instructions due to scarcity of resources or full capacity. The active stater refers to a state of the source entity in which the corresponding source entity is able to receive and fulfil orders effectively within its available capacity and resources. The lesser number of work instructions received by a source entity than a threshold value is referred as the saturation state of the source entity, whereas greater number of work instructions received by the source entity than the threshold value is referred as the active state of the source entity.
The DNO server 104 in the system 100 receives a plurality of incoming work instructions 116 transmitted by the plurality of end-user devices 114 to a plurality of source entities 110 and rearranges the modes of transfer of goods and/or services between the plurality of source entities 110 and the plurality of end-user devices 114 in order to reduce time required in delivery process within the delivery network 102.
In an embodiment, the delivery network 102 is an interconnection between the plurality of end-user devices 114 and the plurality of source entities 110 through wireless communication systems. In an implementation, the delivery network 102 includes a plurality of different geographical areas 108, such as a first geographical area 108A, a second geographical area 108B up to an Nth geographical area 108N. Further, the plurality of source entities 110 are situated in each of the plurality of different geographical areas 108. With reference to FIG. 1, there are shown the plurality of source entities 110, such as a first source entity 110A up to an Nth source entity in the first geographical area 108A, a second source entity 110B up to an Mth source entity in the second geographical area, and a third source entity 110C up to a Pth source entity in the Nth geographical area 108N. Here, the values of N, M, and P are positive integers. Further, the delivery network 102 includes a plurality of rider devices 124, such as a first rider device 124A, a second rider device 124B up to an Nth rider device 124N. The rider refers to a delivery personnel having a delivery vehicle and the rider device refers to a communication device associated with each rider. The rider acts as a mode of transfer for goods and/or services from the plurality of source entities 110 to the plurality of end-user devices 114.
The DNO server 104 is configured to communicate with the delivery network 102 via a communication network 112. In an implementation, the DNO server 104 may be a master server or a master machine that is a part of a data center that controls an array of other cloud servers communicatively coupled to it for load balancing, running customized applications, and efficient data management. Examples of the DNO server 104 may include, but are not limited to a cloud server, an application server, a data server, or an electronic data processing device.
The communication network 112 includes a medium (e.g., a communication channel) through which the plurality of rider devices 124 communicates with the DNO server 104. The communication network 112 may be a wired or wireless communication network. Examples of the communication network 112 may include, but are not limited to, Internet, a Local Area Network (LAN), a wireless personal area network (WPAN), a Wireless Local Area Network (WLAN), a wireless wide area network (WWAN), a cloud network, a Long-Term Evolution (LTE) network, a plain old telephone service (POTS), a Metropolitan Area Network
The hardware processor 106 refers to a computational element that is operable to respond to and processes instructions that drive the system 100. The hardware processor 106 may refer to one or more individual processors, processing devices, and various elements associated with a processing device that may be shared by other processing devices. Additionally, the one or more individual processors, processing devices, and elements are arranged in various architectures for responding to and processing the instructions that drive the system 100. In some implementations, the hardware processors 106 may be an independent unit and may be located outside the DNO server 104 of the system 100. Examples of the hardware processors 106 may include but are not limited to, a hardware processor, a digital signal processor (DSP), a microprocessor, a microcontroller, a complex instruction set computing (CISC) processor, an application-specific integrated circuit (ASIC) processor, a reduced instruction set (RISC) processor, a very long instruction word (VLIW) processor, a state machine, a data processing unit, a graphics processing unit (GPU), and other processors or control circuitry.
The primary storage 106A is configured to store the plurality of incoming work instructions 116 and one or more service domains 118 associated with the plurality of incoming work instructions 116 received from the plurality of end user devices 114. Examples of implementation of the primary storage 106A may include, but are not limited to, an Electrically Erasable Programmable Read-Only Memory (EEPROM), Dynamic Random-Access Memory (DRAM), Random Access Memory (RAM), Read-Only Memory (ROM), Hard Disk Drive (HDD), Flash memory, a Secure Digital (SD) card, Solid-State Drive (SSD), and/or CPU cache memory.
The allocation component 122 is a processing unit, which is configured to allocate the plurality of incoming work instructions 116 to the plurality of rider devices 124 based on an information received through the hardware processor 106. Examples of the allocation component 122 may include but are not limited to, a hardware processor, a digital signal processor (DSP), a microprocessor, a microcontroller, a complex instruction set computing (CISC) processor, an application-specific integrated circuit (ASIC) processor, a reduced instruction set (RISC) processor, a very long instruction word (VLIW) processor, a state machine, a data processing unit, a graphics processing unit (GPU), and other processors or control circuitry.
In operation, the hardware processor 106 is configured to demark a plurality of different geographical areas 108 having the plurality of source entities 110. The demarking refers to a process of dividing or delineating a total geographical area under the delivery network 102 into small geographical areas. The details of the total geographical area within the delivery network 102 is predefined and corresponding geographical coordinates are stored within the primary storage 106A. The purpose of demarking is to establish distinct boundaries or regions within which the plurality of source entities 110 operate. The plurality of demarked geographical areas 108 are typically demarked to group together the plurality of source entities 110 that operate within each respective geographical area. In an implementation, the demarking of the plurality of different geographical areas is performed based on proximity of the plurality of source entities 110 and corresponding geographical coordinates. For example, there are 5 source entities within the delivery network 102. Here, 3 source entities are in close proximity with each other and rest 2 source entities are at large distance from the first 3 source entities, but in close proximity with each other. In such case, the hardware processor 106 demarks a first geographical area with 3 source entities and a second geographical area with 2 source entities. By considering the geographical proximity of source entities for demarking, the plurality of incoming work instructions 116 are allocated to the plurality of rider devices 124 in a manner that minimizes transportation distances and reduces fuel consumption.
After demarking the plurality of different geographical areas 108, the hardware processor 106 is configured to obtain the plurality of incoming work instructions 116 received from the plurality of source entities 110 through the communication network 112 and identify the plurality of incoming work instructions 116 associated with the plurality of source entities 110 from the plurality of end-user devices 114 based on the demarked plurality of different geographical areas 108. The hardware processor 106 matches the plurality of incoming work instructions 116 with the demarked geographical areas 108 to determine the relevant source entities from the plurality of source entities 110 associated with each instruction. For example, if a work instruction originates from an end-user device to a source entity within a demarked geographical area, the hardware processor 106 associates the work instruction with the corresponding geographic area, which ensures that the plurality of incoming work instructions 116 are routed to the source entities in a geographically optimized manner, minimizing travel distances and enhancing delivery efficiency.
The hardware processor 106 determine a frequency of the plurality of incoming work instructions 116 for each of the plurality of source entities 110 within the plurality of different geographical areas 108. The frequency of the plurality of incoming work instructions 116 refers to the amount of the work instructions are received by the hardware processor 106 from the plurality of end-user devices 114 for goods and/or services from the plurality of source entities 110 in a predefined time duration. For example, if one or more end-users are requesting a particular good and/or service from a particular source entity, then number of such requests in an hour defines the frequency of the plurality of incoming work instructions 116 for that particular source entity. Further, the hardware processor 106 is configured to determine one or more service domains 118 of each work instruction from the plurality of incoming work instructions 116. The service domain refers to a specific category or type of good or service for a work instruction. Examples of the service domain may include, but are not limited to, a food item, a grocery item, a medicine, an electronic item, a furniture, a fashion or cosmetic product and the like. The hardware processor 106 receives an incoming work instruction 116 from an end-user device and analyses the work instruction to determine which service domain(s) the work instruction belongs to. Such analysis is based on the content, attributes, or specified parameters within the work instruction. In an implementation, the work instruction includes details such as name of a source entity, details of item for which the work instruction is placed, details of end-user placing the work instruction, details of location of the source entity associated with the work instruction and the like. In an implementation, the hardware processor 106 includes a signal processing module and a natural language processing module, which determines the service domain from the work instruction based on keywords or other predefined rules. Based on the frequency of the plurality of incoming work instructions and corresponding service domains, the hardware processor 106 is further configured to determine the frequency of the incoming work instructions of each domain within each demarked geographical area, which are being placed by the plurality of end-users 110 over a specific period of time.
After determining the frequency of the plurality of work instructions for each service domain, the hardware processor 106 is configured to determine a trend in the frequency of the plurality of work instructions for each service domain within the specific period of time.
For example, consider a week-long time period for trend identification and three source entities A, B, and C. The source entities A, B, and C belongs to different geographical areas. In the first few days of the week, the source entity A receives a moderate frequency of incoming work instructions belonging to a first service domain, such as an average of 10 work instructions per day. In the middle of the week, there is a sudden surge in the frequency of work instructions belonging to a second service domain for source entity B, such as an average of 20 work instructions per day. On the last couple of days of the week, the source entity C experiences a decline in the frequency of work instructions belonging to a third service domain, such as an average of 5 work instructions per day. Based on the above example, the trend in the frequency of incoming work instructions can be identified as follows:
Source Entity A: stable trend - the frequency of the plurality of work instructions belonging to the first service domain remains consistent throughout the week, indicating a stable workload for the source entity A.
Source Entity B: increasing trend - there is a significant increase in the frequency of work instructions belonging to the second service domain during the middle of the week, suggesting an increasing workload for the source entity B.
Source Entity C: decreasing trend - the frequency of work instructions belonging to the third service domain decreases towards the end of the week, indicating a decreasing workload for the source entity C.
Further, the hardware processor 106 is configured to determine whether one or more source entities from the plurality of source entities 110 are in a saturation state or an active state 120 in an upcoming time period based on the trend in the frequency of the plurality of incoming work instructions 116 received from the plurality of end-user devices 114 and the one or more determined service domains 118. At the saturation state, the source entity is unable to receive or fulfil work instructions due to scarcity of resources or full capacity. At the active state, the source entity is able to receive and fulfil orders effectively within its available capacity and resources. The hardware processor 106 is configured to collect a historical data of the trend in the frequency of the plurality of incoming work instructions 116 related to the one or more service domains 118 for each source entity within a specific time period and predicts the saturation state or the active state of the plurality of source entities 110 in the upcoming time period. The determination of the saturation state and active state is explained with the help of following example:
The primary storage 106A is stored with the historical data that includes the frequency of incoming work instructions and the one or more service domains associated with them such as: a restaurant A: Frequency - 10 work instructions/hour, Service Domain - Food Delivery, a grocery Store B: Frequency - 15 work instructions/hour, Service Domain - Grocery Delivery, a retail Shop C: Frequency - 25 work instructions/hour, Service Domain - Retail Delivery.
Trend Analysis: The hardware processor 106 analyses the historical data to identify trends in the frequency of work instructions for each service domain, which is indicative of how the workload varies over time. Assume the following trends: food Delivery for restaurant A: Increasing trend in the frequency of work instructions, grocery Delivery for grocery store B: Fluctuating trend in the frequency of work instructions, and retail Delivery for retail shop C: Decreasing trend in the frequency of work instructions.
Projection to Upcoming Time Period: Based on the observed trends, the hardware processor 106 predicts the frequency of incoming work instructions for each service domain in an upcoming time period, such as the next hour. For example, the food Delivery for restaurant A: projected frequency - 12 work instructions/hour, the grocery delivery for grocery store B: projected frequency - 18 work instructions/hour, the retail delivery retail shop C: projected frequency - 20 work instructions/hour.
Saturation and Active State Determination: using the trend in the frequency of incoming work instructions, the hardware processor 106 is configured to determine the saturation state and the active state for each source entity in a demarked geographical area in the upcoming time period.
Saturation State: If the determined frequency of incoming work instructions reduces below a predefined saturation threshold specific to each service domain and source entity, a saturation state gets defined for the source entity. For example, if the saturation threshold for food delivery is set at 15 work instructions/hour, then restaurant A would be in a saturation state in the upcoming time period.
Active State: if the predicted frequency of incoming work instructions exceeds the saturation threshold, then the source entity is determined to be in an active state. For example, if the saturation threshold for Grocery Delivery is set at 15 work instructions/hour, then Grocery Store B would be in an active state in the upcoming time period.
The system 100 further includes the allocation component 122 which is configured to automatically allocate the plurality of incoming work instructions 116 to a plurality of rider devices 124 based on the saturation state or the active state of the plurality of source entities 110, the frequency of the plurality of incoming work instructions 116 and the determined one or more service domains 118. In an implementation, the hardware processor 106 determines the saturation state and active state of the source entities within each demarked geographical area. For example, a geographical area A has 10 source entities, where 8 of them will be in saturation state and 2 will be in active state in next 2 hours. Further, there is another geographical area B, which is also having 12 source entities, where 9 of them will be in active state and 3 will be in saturation state in next 2 hours. In addition, there are 20 riders available in the delivery network 102. Based on such information, the allocation component 122 allocates 15 riders to serve the source entities in the geographical area B and 5 riders to the source entities in the geographical area A (as the geographical area B is having a greater number of active source entities as compared to the geographical area A). Further, the allocation component 122 directs the 15 riders to be present at the geographical area B due to more number active source entities being functional for next 2 hours. Due to allocation of more riders to the geographical area B than the geographical area A, the fulfilment of work instructions in the geographical area B will be done rapidly by avoiding delay in the delivery process and avoid idle time for the plurality of rider devices 124.
In another example, consider the saturation state and the active state on the basis of the one or more service domains 118. The delivery network 102 includes two geographical areas C and D. Based on the frequency of the plurality of incoming work instructions 116 and the one or more service domains 118, the hardware processor 106 estimates that the maximum source entities in the geographical area A will be in active state for work instructions related to grocery items and the maximum source entities in the geographical area A will be in active state for work instructions related to food items. In such case, the allocation component 122 allocates the work instructions received by the source entities in the geographical area A to a group of riders having historical data of delivering grocery items. Similarly, the allocation component 122 allocates the work instructions received by the source entities in the geographical area B to another group of riders having historical data of delivering food items. Such rider allocation is beneficial for increasing operational efficiency and optimized delivery network 102.
In accordance with an embodiment, the saturation state, or the active state of the plurality of source entities 110 in the upcoming time period is determined through a semi-supervised machine learning model. The implementation of the semi-supervised machine learning model is explained through following example –
Data collection: the hardware processor 106 is configured to analyse the historical data on the frequency of incoming work instructions for each source entity, along with the corresponding service domains. For instance, source entities in a geographical area A received an average of 50 work instructions per day in the food delivery service domain, while source entities in a geographical area B received 30 work instructions per day in the electronics delivery service domain.
Feature extraction: from the historical data, the hardware processor 106 is configured to extract relevant features such as the frequency of work instructions and the service domains associated with each entity.
Model training: here, the hardware processor 106 is configured to implement the semi-supervised model, to train on the historical data. A subset of the historical data may be labelled to indicate the saturation state or active state for certain source entities. For example, the source entities in the geographical area A are labelled as being in a saturation state during busy periods based on the historical data.
Pattern recognition: the trained model recognizes patterns or associations between the extracted features and the saturation or active state of the source entities. The model learns that low work instruction frequencies in certain service domains may indicate a potential saturation state, while higher frequencies suggest an active state.
State determination: when new data is provided as input, such as the projected frequency of work instructions and the determined service domains for the upcoming time period, the model applies its learned patterns. For example, based on the projected work instruction frequency of 60 per day for Entity A in the food delivery service domain, the model predicts that Entity A might be in an active state due to the increased workload.
Decision making and optimization: the determined saturation and active states of the source entities guide decision-making in the delivery network 102. For instance, knowing that the source entity in the geographical area A are likely to be in a saturation state, the hardware processor 106 allocates the work instructions associated with the geographical area A to the lesser number of riders than other active source entities in other geographical areas. By using the semi-supervised machine learning model, the hardware processor can proactively identify potential saturation or active states of the source entities based on the projected workload and service domains, which allows for optimized resource allocation, workload management, decision-making, and ultimately enhancing the efficiency and effectiveness of the delivery network 102.
In accordance with an embodiment, the hardware processor 106 is further configured to generate notifications or alerts for the one or more source entities from the plurality of source entities 110 that are determined to be in the saturation state in the upcoming time period. In an implementation, the hardware processor 106 is configured to transmit the notifications or alerts to corresponding electronic devices associated with the plurality of source entities 110. In an implementation, the notification or alert is in the form of a text message. The notification or alert of the saturation state and the active state is also beneficial for source entities to determine which type of items have to be produced for an upcoming predefined time period. For example – the hardware processor 106 demarks the geographical area based on the types of source entities, such as restaurant, grocery, and pharmacy as geographical areas A, B and C respectively. Further, the hardware processor 106 determines that the source entities in the geographical area A will be in the active state for “paneer” food items and will be in saturation state for “Chinese” food items. The hardware processor 106 transmits the notification or alert to the source entities in the geographical area A to increase the preparation of the paneer food items and reduce the preparation of the Chinese food items for a particular time period in order to increase sales and profit of the source entities.
In accordance with an embodiment, the hardware processor 106 is further configured to generate a plurality of optimized delivery routes for the plurality of rider devices 124 based on the plurality of demarked geographical areas 108 and the determined saturation states and active states of the plurality of source entities 110. The optimized delivery route refers to a strategically planned path that enables efficient and timely delivery of goods or services. The hardware processor 106 takes into consideration the location of the source entities within each geographical area and their corresponding states. The generation of optimized delivery route is explained with the help of following example: The hardware processor 106 demarks the geographical areas, such as area A, area B, area C, and area D. In an example, the hardware processor 106 determines active states for source entities in the areas A and D, whereas determines saturation states for source entities in the areas B and C. Further, the hardware processor 106 is configured to implement a route planning module to generate optimized delivery routes by taking account of factors such as distance between the geographical areas, the states of source entities in each geographical area, and the capacity of the plurality of riders. In an implementation, the route planning module is a processing unit. Further, the route planning module assigns priorities to the delivery routes by providing a higher priority to routes in areas with active source entities (such as area A and D) and lower priority to routes in the areas with saturated source entities (such as area B and C). The hardware processor 106 optimizes the delivery routes to minimize travel distance and time, which helps the plurality of riders to complete their work instructions efficiently. The generated delivery routes aim to ensure efficient and effective fulfilment of the plurality of incoming work instructions 116. For example, if a particular geographical area has multiple source entities in an active state, the processor can assign rider devices from other areas with saturated source entities to balance the workload and avoid delays.
In accordance with an embodiment, the allocation component 106 is further configured to re-balance the delivery network 102 by adjusting the allocation of the plurality of incoming work instructions 116 to the plurality of rider devices 124 based on real-time or near real time changes in: the frequency of the plurality of incoming work instructions 116, the one or more service domains 118 of the plurality of incoming work instructions 116; and the saturation and active state of the plurality of source entities 110 to optimize the delivery network 102. The real time changes refer to immediate or instantaneous updates in the frequency of the plurality of incoming work instructions 116, service domains 118, and the saturation and active state of the plurality of source entities 110. The near real time changes refer to updates in the frequency of the plurality of incoming work instructions 116, service domains 118, and the saturation and active state of the plurality of source entities 110 that occur with minimal delay, typically within a short timeframe. The re-balancing of the delivery network 102 is explained with the help of following scenario:
In real-time, as new work instructions are placed by the plurality of end-user devices 114 through the DNO server 104, the allocation component 122 immediately receives the information about the incoming work instructions through the hardware processor 106. The hardware processor 106 analyses the workload distribution, the service domains (such as different types of cuisines), and the availability of source entities (restaurants and kitchens) within the delivery network 102. Based on the real-time changes, the hardware processor 106 directs the allocation component to dynamically allocate the work instructions to the most suitable rider devices for efficient and timely delivery. For example, if there is a sudden surge in the work instructions related to food orders during a busy lunch hour, the hardware processor 106 identifies the increased workload from the frequency of the plurality of work instructions and directs the allocation component 122 to dynamically allocate more work instructions to available riders and ensure prompt deliveries.
The system 100 determines the saturation state or active state of the source entities in demarked geographical areas in advance and allocates the riders to serve the demarked geographical areas in proportion to the number of active and saturated source entities (more allocation of the riders to active areas than saturated areas), thereby provide balanced distribution of workload among the plurality of riders and reducing delay in the delivery process. The system 100 changes the work instructions allocated to the plurality of rider devices 124 in accordance with change in saturation and active states of the source entities, which enables the system 100 to provide timely fulfilment of the work instructions from the source entities having dynamic fluctuation in operational state. The system 100 provides notification or alert to the plurality of source entities 110 regarding upcoming saturation or active state, thereby enabling the plurality of source entities 110 to manage corresponding operations to cope up with increasing (active state) or decreasing (saturation state) demand. The system 100 provides optimized delivery routes in accordance with the saturation and active states and prioritize the delivery routes in geographical areas with active source entities to minimize travel distance and time in the delivery process. By using the semi-supervised machine learning model, the system 100 proactively identifies the potential saturation or active states of the source entities based on the projected workload and service domains, which allows for optimized resource allocation, workload management, decision-making, and ultimately enhancing the efficiency and effectiveness of the delivery network 102.
Additionally, the system 100 achieves following technical effects and advantage over conventional delivery network management systems.
a) The ability of the system 100 to demarcate different geographical areas within the delivery network 102 allows for better organization and management of the plurality of incoming work instructions 116. By defining specific regions, the system 100 can streamline the allocation process and ensure that tasks are assigned to the riders within their designated service domains.
b) Based on the trend in the frequency of incoming work instructions 116 and the determined service domains 118, the system 100 can proactively assess whether source entities are going to be in a saturation state or an active state in the upcoming time period. This assessment allows the system 100 to understand the workload capacity of each source entity and make allocation decisions beforehand accordingly. By considering the saturation and activity levels, the system 100 can avoid overburdening source entities or underutilizing available resources.
c) The system 100 takes into account geographical area demarcation, work instruction frequency analysis, saturation state and active state determination for automated task allocation, which leads to improved allocation efficiency, optimal resource utilization, and adaptability in the delivery network 102 optimization process.
FIG. 2 is a block diagram of a delivery network optimizing (DNO) server in a system for optimizing a delivery network, in accordance with another embodiment of the present disclosure. FIG. 2 is described in conjunction with elements of FIG. 1. With reference to FIG. 2, there is shown a block diagram of a delivery network optimizing (DNO) server 202 in the system 100 (of FIG. 1) for optimizing the delivery network 102 (of FIG. 1). The DNO server 202 includes a network interface 204 communicatively coupled to the hardware processor 106 (of FIG. 1). Further, the DNO server 202 includes a primary storage 206 and the allocation component 108 communicatively coupled with the hardware processor 106.
The network interface 204 refers to a communication interface to enable communication of the DNO server 202 to any other external device, such as a user device. Examples of the network interface 204 include, but are not limited to, a network interface card, a transceiver, and the like.
The primary storage 206 is configured to store a profile information regarding the plurality of riders operating the plurality of rider devices 124. The profile information includes one or more datasets, such as a historical incoming work instructions dataset 208, a behavioural dataset 210, and a vehicle dataset 212. The historical incoming work instructions dataset 208 contains a historical data of previous work instructions fulfilled by each rider from the plurality of riders operating the plurality of rider devices 124. In an implementation, the historical incoming work instructions dataset 208 includes information such number of work instructions fulfilled by each rider in past time period, number of work instructions refused by each rider in the past time period and the like.
The behavioural dataset 210 includes an information about the behaviour and performance of the plurality of riders within the delivery network 102. The behavioural dataset 210 includes the information such as delivery speed, customer ratings, adherence to delivery schedules and the like for each rider from the plurality of riders in the delivery network 102. The vehicle dataset 212 includes an information about the delivery vehicles assigned to each rider device. The vehicle dataset 212 the information such as vehicle type, capacity, availability, mileage of the vehicle and the like.
In accordance with an embodiment, the allocation of the plurality of incoming work instructions 116 to the plurality of rider devices 124 by the allocation component 122 is further based on the historical incoming work instructions dataset 208, the behavioural dataset 210, and the vehicle dataset 212 associated with a plurality of delivery vehicles assigned to each rider device from the plurality of rider devices 124. The allocation component 122 is configured to allocate work instructions based on the past performance, the rider behaviour, the vehicle suitability, and the workload distribution to optimize the allocation process. For example, if a rider has consistently demonstrated fast delivery times and high customer ratings in the behavioural dataset, then the allocation component 122 prioritizes the assignment of the time-sensitive or high-priority work instructions to that rider. Similarly, if a particular vehicle associated with a rider device has a larger capacity, then the allocation component 122 assigns bulk work instructions or larger deliveries to optimize the rider’s vehicle usage. By incorporating the historical incoming work instructions dataset 208, the behavioural dataset 210 and the vehicle dataset 212 into the allocation process, the hardware processor 106 makes intelligent decisions while assigning work instructions to the plurality of riders, which helps in optimizing resource utilization, improve delivery efficiency, and enhance overall customer satisfaction.
FIG. 3 is a block diagram of a system for optimizing a delivery network, in accordance with another embodiment of the present disclosure. With reference to FIG. 3, there is shown a block diagram of a system 300 for optimizing the delivery network 102 (of FIG. 1), which includes the DNO server 104 having the hardware processor 106, the allocation component 122 and the primary storage 106A communicatively coupled to each other. The primary storage 106A is configured to store the plurality of incoming work instructions 116 and the one or more service domains 118 received from the plurality of end user devices 114 (of FIG. 1). Further, the DNO server 104 is connected with the delivery network 102 through the communication network 112.
In accordance with an embodiment, the hardware processor 106 is configured to determine a density of work area hotspots 304 in each of the demarked geographical areas from the plurality of different geographical areas 108 based on the determined frequency of the plurality of incoming work instructions 116 and the one or more service domains 118 to rebalance the delivery network 102. In a scenario, the delivery network 102 includes the plurality of demarked geographical areas, such as a first geographical area 108A, a second geographical area 108B up to an nth geographical area 108N. The hardware processor 106 is configured to determine the work area hotspots 304, such as a first work area hotspot 304A up to N1th work area hotspot 304N1 in the first geographical area 108A, a second work area hotspot 304B up to N2th work area hotspot 304N2 in the second geographical area 108B, a third work area hotspot 304C up to Nth work area hotspot 304N in the nth geographical area 108N. The work area hotspots refer to locations or zones with a high concentration of work instructions of definite service domains. In accordance with an embodiment, the work area hotspots 304 include one or more source entities from the plurality of source entities 110 having the frequency of work instructions above a pre-defined threshold. In an implementation, the primary storage 106A of the system 300 is stored with the predefined threshold of the frequency of the plurality of incoming work instructions 116. Based on the determined frequency of work instructions and the associated service domains, the hardware processor 106 estimates the density of work area hotspots within each geographical area. The hardware processor 106 defines source entities as hotspots, when the frequency of the plurality of incoming work instructions 116 is more than the predefined threshold value. The density of work area hotspots helps to identify geographical areas that experience a heavier workload or demand. With such information, the hardware processor 106 rebalances the allocation of the plurality of work instructions 116 to the plurality of rider devices 124 to effectively handle the workload in each area. For instance, the hardware processor 106 allocates more riders to the areas having high density of work hotspots to ensure efficient and timely fulfilment of the plurality of work instructions 116.
In accordance with an embodiment, the hardware processor 106 is further configured to generate delivery network optimising insights 302 by summarizing the workload distribution, resource allocation, and performance of the delivery network 102 based on the determined density of the work area hotspots 304 . The delivery network optimising insights 302 refers to an information and/or recommendation generated by the hardware processor 106 to enhance the operations of the delivery network 102. In an example, the hardware processor 106 generates the delivery network optimising insights 302, which indicates that certain geographical areas have a high density of work area hotspots indicating a heavy workload. Based on such information, the hardware processor 106 further allocates additional riders or adjust the routing of the plurality of rider devices 124 to address the increased workload and ensure timely fulfilment of work instructions. The generated delivery network optimising insights 302 provide a guidance for enhancing the network's operations and achieving better efficiency and customer satisfaction.
FIG. 4 is a diagram of an exemplary scenario of the allocation of work instructions to a plurality of riders based on saturation state or active state of the source entities, in accordance with another embodiment of the present disclosure. FIG. 4 is described in conjunction with elements of FIG. 1, 2 and 3. With reference to FIG. 4, there are shown three demarked geographical areas, such as a first geographical area 108A, a second geographical area 108B and a third geographical area 108C. The first geographical area 108A includes two source entities, such as a first source entity 110A and a second source entity 110B. Similarly, the second geographical area 108B includes a third source entity 110C and a fourth source entity 110D and the third geographical area 108C includes a fifth source entity 110E and a sixth source entity 110F. Further, the delivery network 102 includes four rider devices, such as a first rider device 124A, a second rider device 124B, a third rider device 124C and a fourth rider device 124D.
In such scenario, the DNO server 104 is configured to determine that the first source entity 110A, the second source entity 110B and the third source entity 110C will be in active state and the fourth source entity 110D, the fifth source entity 110E and the sixth source entity 110F will be in saturation state in next 2 hours. Based on the information related to the saturation state and active state, the DNO server 104 estimates that the geographical area A has more density of active source entities than the geographical area B and the geographical area C. Therefore, the DNO server 104 is configured to allocate the work instructions related to the source entities in the geographical area A to riders with the first rider device 124A and the second rider device 124B. Further, the DNO server 104 is configured to allocate the work instructions related to the source entities in the geographical area B to the rider with the third rider device 124C and in the geographical area C to the rider with the fourth rider device 124D. In other words, the DNO server 104 allocates the plurality of incoming work instructions 116 to the plurality of rider devices 124 based on density of active source entities. In other words, more the density of active source entities in the geographical area, more rider devices are allocated to fulfil the work instructions in the corresponding geographical area.
FIG. 5 is a flowchart of a method for optimizing a delivery network, in accordance with an embodiment of the present disclosure. FIG. 5 is described in conjunction with elements of FIG. 1 to 4. With reference to FIG. 5, there is shown a method 500 for optimizing the delivery network 102 (of FIG. 1). The method 500 includes steps from 502 to 512.
At step 502, the method 500 includes demarking the plurality of different geographical areas 108 having the plurality of source entities 110 in the delivery network optimising (DNO) server 104 (of FIG. 1). The demarking refers to a process of dividing or delineating a total geographical area under the delivery network 102 into small geographical areas. The purpose of demarking is to establish distinct boundaries or regions within which the plurality of source entities 110 operate.
At step 504, the method 500 includes identifying the plurality of incoming work instructions 116 associated with the plurality of source entities 110 from the plurality of end-user devices based on the demarked plurality of different geographical areas 108. The hardware processor 106 matches the plurality of incoming work instructions 116 with the demarked geographical areas 108 to determine the relevant source entities from the plurality of source entities 110 associated with each work instruction.
At step 506, the method 500 includes determining a frequency of the plurality of incoming work instructions for each of the plurality of source entities 110. The frequency of the plurality of incoming work instructions 116 refers to the amount of the work instructions received by the hardware processor 106 from the plurality of end-user devices 114 for goods and/or services from the plurality of source entities 110 in a predefined time duration.
At step 508, the method 500 includes determining one or more service domains of each work instruction from the plurality of incoming work instructions. The service domain refers to a specific category or type of good or service for a work instruction. Examples of the service domain may include, but are not limited to, a food item, a grocery item, a medicine, an electronic item, a furniture item, a fashion, or cosmetic product and the like.
At step 510, the method 500 includes determining whether one or more source entities from the plurality of source entities 110 are in a saturation state or an active state in an upcoming time period based on a trend in the frequency of the plurality of incoming work instructions received from the plurality of end-user devices and the determined one or more service domains. At the saturation state, the source entity is unable to receive or fulfil work instructions due to scarcity of resources or full capacity. At the active state, the source entity is able to receive and fulfil orders effectively within corresponding available capacity and resources. The method 500 includes collecting a historical data of the trend in the frequency of the plurality of incoming work instructions 116 related to the one or more service domains 118 for each source entity within a specific time period and predicts the saturation state or the active state of the plurality of source entities 110 in the upcoming time period.
At step 512, the method 500 includes automatically allocating the plurality of incoming work instructions 116 to a plurality of rider devices 124 based on the saturation state or active state of the plurality of source entities 110, the frequency of the plurality of incoming work instructions 116 and the determined one or more service domains 118. In an implementation, the method 500 includes determining the saturation state and active state of the source entities within each demarked geographical area and allocating the work instructions to the plurality of rider devices 124 in proportion to the active and saturated source entities. In other words, the method 500 includes allocating more rider devices to fulfil the work instructions in the geographical areas having a greater number of active source entities (i.e., a smaller number of saturated source entities than active source entities) than the geographical areas having a smaller number of active source entities (i.e., a greater number of saturated source entities than the active source entities).
The method 500 includes determining the saturation state or active state of the source entities in demarked geographical areas in the upcoming time period and allocates the plurality of incoming work instructions 116 to the number of riders, which are in proportion to the number of active source entities, thereby reducing delay in the delivery process and provide balanced distribution of workload among the plurality of rider devices 124. The method 500 includes changing the plurality of work instructions allocated to the plurality of rider devices 124 in accordance with change in saturation and active states of the plurality of source entities 110, which enables timely fulfilment of the work instructions from the source entities having dynamic fluctuation in operational state. The determination of the upcoming saturation or active state of source entities based on trends in the method 500 allows for proactive decision-making regarding workload distribution. By predicting the workload and availability of source entities in advance, the method 500 includes making informed decisions to balance the workload across the delivery network 102, preventing bottlenecks and ensuring timely deliveries.
Additionally, the method 500 achieves following technical effects and advantage over conventional delivery network management systems.
a) The ability of the DNO server 104 used in the method 500 to demarcate different geographical areas within the delivery network 102 allows for better organization and management of the plurality of incoming work instructions 116. By defining specific regions, the DNO server 104 in the method 500 can streamline the allocation process and ensure that tasks are assigned to the riders within their designated service domains.
b) Based on the trend in the frequency of incoming work instructions 116 and the determined service domains 118, the DNO server 104 in the system 100 can proactively assess whether source entities are going to be in a saturation state or an active state in the upcoming time period. This assessment allows the DNO server 104 in the method 500 to understand the workload capacity of each source entity and make allocation decisions beforehand accordingly. By considering the saturation and activity levels, the DNO server 104 in the method 500 can avoid overburdening source entities or underutilizing available resources.
c) The DNO server 104 in the method 500 takes into account geographical area demarcation, work instruction frequency analysis, saturation state and active state determination for automated task allocation, which leads to improved allocation efficiency, optimal resource utilization, and adaptability in the delivery network 102 optimization process.
The steps 502 to 512 are only illustrative and other alternatives can also be provided where one or more steps are added, one or more steps are removed, or one or more steps are provided in a different sequence without departing from the scope of the claims herein. Various embodiments and variants disclosed with the aforementioned system (such as the system 100) apply mutatis mutandis to the aforementioned method 500.
Modifications to embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as "including", "comprising", "incorporating", "have", "is" used to describe and claim the present disclosure are intended to be construed in a non-exclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural. The word "exemplary" is used herein to mean "serving as an example, instance or illustration". Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and/or to exclude the incorporation of features from other embodiments. The word "optionally" is used herein to mean "is provided in some embodiments and not provided in other embodiments". It is appreciated that certain features of the present disclosure, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the present disclosure, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable combination or as suitable in any other described embodiment of the disclosure.
, Claims:CLAIMS
1. A system (100, 300) for optimizing a delivery network (102), the system (100) comprises:
a delivery network optimising (DNO) server (104) comprising a hardware processor (106) configured to:
demark a plurality of different geographical areas (108) having a plurality of source entities (110);
identify a plurality of incoming work instructions (116) associated with the plurality of source entities (110) from a plurality of end-user devices (114) based on the demarked plurality of different geographical areas (108);
determine a frequency of the plurality of incoming work instructions (116) for each of the plurality of source entities (110);
determine one or more service domains (118) of each work instruction from the plurality of incoming work instructions (116);
determine whether one or more source entities from the plurality of source entities (110) are in a saturation state or an active state (120) in an upcoming time period based on a trend in the frequency of the plurality of incoming work instructions (116) received from the plurality of end-user devices (114) and the one or more determined service domains (118); and
an allocation component (122) configured to automatically allocate the plurality of incoming work instructions (116) to a plurality of rider devices (124) based on the saturation state or active state of the plurality of source entities (110), the frequency of the plurality of incoming work instructions (116) and the determined one or more service domains (118).
2. The system (100, 300) as claimed in claim 1, wherein allocation of the plurality of incoming work instructions (116) to the plurality of rider devices (124) by the allocation component (122) is further based on historical incoming work instructions dataset (208), behavioural dataset (210), and vehicle dataset (212) associated with a plurality of delivery vehicles assigned to each rider device from the plurality of rider devices (124).

3. The system (100, 300) as claimed in claim 1, wherein the saturation state or the active state of the plurality of source entities (110) in the upcoming time period is determined through a semi-supervised machine learning model.
4. The system (100, 300) as claimed in claim 1, wherein the hardware processor (106) is further configured to generate a plurality of optimized delivery routes for the plurality of rider devices (124) based on the plurality of demarked geographical areas (108) and the determined saturation states and active states of the plurality of source entities (110).
5. The system (100, 300) as claimed in claim 1, wherein the hardware processor (106) is further configured to generate notifications or alerts for the one or more source entities from the plurality of source entities (110) that are determined to be in the saturation state in the upcoming time period.
6. The system (100, 300) as claimed in claim 1, wherein the allocation component (106) is further configured to re-balance the delivery network (102) by adjusting the allocation of the plurality of incoming work instructions (116) to the plurality of rider devices (124) based on real-time or near real time changes in:
the frequency of the plurality of incoming work instructions (116);
one or more service domains (118) of the plurality of incoming work instructions (116); and
saturation and active state of the plurality of source entities (110) to optimize the delivery network (102).
7. The system (300) as claimed in claim 1, wherein the hardware processor (106) is configured to determine a density of work area hotspots (304) in each of the demarked geographical areas from the plurality of different geographical areas (108) based on the determined frequency of the plurality of incoming work instructions (116) and the one or more service domains (118) to rebalance the delivery network (102).
8. The system (300) as claimed in claim 6, wherein the hardware processor (106) is further configured to generate delivery network optimising insights (302) summarizing the workload distribution, resource allocation, and performance of the delivery network (102) based on the determined density of the work area hotspots (304).
9. The system (300) as claimed in claim 6, wherein the work area hotspots (304) comprise one or more source entities from the plurality of source entities (110) having the frequency of the plurality of work instructions (116) above a pre-defined threshold.
10. A method (200) for optimizing a delivery network (102), the method (200) comprises:
in a delivery network optimising (DNO) server (104):
demarking a plurality of different geographical areas (108) having a plurality of source entities (110);
identifying a plurality of incoming work instructions (116) associated with the plurality of source entities (110) from a plurality of end-user devices (114) based on the demarked plurality of different geographical areas;
determining a frequency of the plurality of incoming work instructions (116) for each of the plurality of source entities (110);
determining one or more service domains (118) of each work instruction from the plurality of incoming work instructions (116);
determining whether one or more source entities from the plurality of source entities (110) are in a saturation state or an active state in an upcoming time period based on a trend in the frequency of the plurality of incoming work instructions (116) received from the plurality of end-user devices (114) and the determined one or more service domains (118); and
automatically allocating the plurality of incoming work instructions (116) to a plurality of rider devices (124) based on the saturation state or active state of the plurality of source entities (110), the frequency of the plurality of incoming work instructions (116) and the determined one or more service domains (118).