TWI769928B - System reliability evaluation method for food online delivery system - Google Patents

Abstract

The food online delivery system can be constructed as a multistate food online delivery network (MFODN), and use corresponding capacity probability table for different units of time to calculate system's reliability. Courier's deliveries will be restricted by their delivery box’s space. Therefore, the space constraint during the delivery process should be considered. In addition, customer satisfaction that has a strong positive correlation with delivery time. Customer satisfaction can be improved while the delivery arrives on time, therefore the couriers delivering the orders within the given time threshold is crucial. According an algorithm proposed in this invention, the performance of MFODN can be calculated, and the system reliability can be evaluated. Based on the proposed system reliability, managers can know the MFODN’s ability to fulfill the given set of orders and make appropriate decisions if necessary.

Description

System reliability evaluation method of catering online distribution system

The invention relates to the field of computing technology, in particular to a system reliability evaluation method for an online catering distribution system.

With the development of Internet technology, popularization of smart phones, busy people, and changes in lifestyle, the online food delivery (OFD) market is booming. Service quality is an important factor in gaining a competitive advantage in the Food & Beverage industry. Online Food Delivery (OFD) is an industry dedicated to assisting customers in providing food delivery services. In recent years, as the industry continues to flourish, its service quality has also attracted the attention of many consumers.

The difference between online food delivery system and general logistics is its instant and diversity. Immediacy means that the items of the online food delivery system are generated in real-time. The delivery service must start as soon as the customer makes the item. The characteristic of diversity comes from the diversity of objects. That is, it typically transports different kinds of items, rather than similar items as in general logistics.

Since food delivery personnel can choose their work location through the online food delivery system, the number of available food delivery personnel in each food delivery area is random. To evaluate the performance of systems when randomness is involved, many researchers use network analysis methods to transform different systems into network topologies, including power networks, manufacturing networks, transportation networks, and computer networks. The literature defines the capacity on transmission edges (arcs) or nodes (nodes) as random, and calls such a network with random capacity a multi-state network.

For the system performance evaluation of the above-mentioned multi-state network, the existing literature focuses on the construction of an optimized mathematical model, which aims to minimize the cost under the constraints of the delivery time, and find the best solution for different needs. The best distribution plan, or analyze the relationship between service indicators and customer satisfaction with the service received, so as to determine the degree of service quality that each factor affects customer perception, or use quantitative methods to construct a distribution service quality table as a basis for evaluating service levels, and include it in Management considerations to control for the more influential factors. Or for the probability of fulfilling the distribution demand of the integrated transportation and logistics system, the system performance index of the transportation and logistics system under the time threshold limit is discussed.

However, the above literature only considers the customer's service evaluation of the distribution system after the fact, and regards it as a quantitative index of management management, and does not consider the factors that may change.

Although the online food delivery system is one of the emerging transportation systems and has been proven to bring convenience to people's lives, the literature or patents related to the online food delivery system are quite rare, especially how to quantify the load of the system related research.

To sum up, in view of the fact that the existing literature and related patents related to the online catering distribution system do not consider the probabilistic factor, the present invention takes into account the probabilistic characteristics of the different number of delivery personnel per unit time in each area in practical situations. , according to the historical data to evaluate the probability distribution of the distribution personnel status in each region in the catering online distribution system, and then calculate the system reliability over time, which can provide managers as a performance indicator of the management system.

The invention defines the system reliability as the probability that the distribution personnel in the system can successfully deliver all orders under a given space and time threshold. Compared with other documents or patents, the system reliability evaluation method of the online catering distribution system proposed by the present invention can see the real-time online catering distribution system status, and this indicator can also be provided to the relevant distribution companies to understand the system distribution personnel. Whether the distribution satisfies the order basis for the assignment or management of delivery personnel.

According to the above description, the present invention proposes to solve the deficiencies existing in the prior art, and proposes A system reliability evaluation method for an online catering distribution system, the method comprising the following steps: constructing a network model by using a network analysis method and inputting parameters such as time threshold, space limitation of delivery personnel, restaurant waiting time, and delivery time of each route; Calculate the minimum number of the delivery personnel that meets the space limit; use the branch-and-bound method to find the order delivery combination that meets the space limit; The conditions are converted into a least lower bound vector; and the reliability of the system is calculated.

In one embodiment, the above-mentioned network model converts the catering online distribution system into a network topology considering the randomness of the above-mentioned delivery personnel.

In one embodiment, the network topology includes a plurality of nodes, a plurality of transmission edges connecting the plurality of nodes, a commodity, a spatial weight of the commodity, a waiting time of the commodity, and a travel time to transmit the commodity.

In one embodiment, the capacity of each node in the plurality of nodes is a random variable, and the probability thereof is based on a given probability distribution.

In one embodiment, the above-mentioned use of the branch-and-bound method to find the order delivery combination that satisfies the space limitation is to arrange the commodities in the order according to the weight of the commodities in descending order, and then use the branch-and-bound method to allocate the delivery of the commodities in the order, After allocating the delivery of the commodities in the order using the above branch and bound method, a first-level solution for commodity allocation is obtained.

In one embodiment, each of the first-level solutions for the commodity allocation obtains different second-level solutions for the commodity allocation through permutation.

In one embodiment, the reliability of the system is evaluated using a recursive sum of disjoint products (RSDP) algorithm.

From another aspect of the present invention, the present invention proposes a system reliability evaluation method for an online catering distribution system, the method comprising the following steps: constructing a network model and inputting a time threshold, Parameters such as space limitation of delivery personnel, restaurant waiting time, delivery time of each route, etc.; calculate the minimum number of delivery personnel that meet the space limitation; use the branch-and-bound method to find the order delivery combination that meets the space limitation. The weights of the commodities are arranged from large to small, and then the distribution of the commodities in the order is allocated by the branch and bound method. After the distribution of the commodities in the order is allocated by the above branch and bound method, the first-level solution about commodity allocation is obtained; using the commodity The first-level solution of the distribution generates a second-level solution for the distribution of the commodity through permutation; calculate the transportation time of the second-level solution for the distribution of the commodity above, and delete the second-level solution that is not feasible; use the above-mentioned about The second-level solution of the commodity allocation obtains the capacity vector; the reliability of the system is calculated.

In one embodiment, the above-mentioned network model converts the catering online distribution system into a network topology considering the randomness of the above-mentioned delivery personnel.

In one embodiment, the network topology includes a plurality of nodes, a plurality of transmission edges connecting the plurality of nodes, a commodity, a spatial weight of the commodity, a waiting time of the commodity, and a travel time to transmit the commodity.

In one embodiment, calculating the transportation time of the second-level solution of the commodity distribution, and deleting the infeasible second-level solution further includes the following steps: creating two sets to store the feasible and infeasible commodities respectively Allocate; list each courier in the second level solution and calculate the transit time; and list all feasible second level solutions.

In one embodiment, the reliability of the system is evaluated using a recursive sum of disjoint products (RSDP) algorithm.

100: Multi-Order State Networks

201, 203, 205, 207, 209, 211: Steps

300: Devices for predicting system reliability

301: Input device

303: memory

305: Processor

307: Output device

3031: Algorithms

3032: SODS network topology

FIG. 1 shows a schematic diagram of a multi-level state network of the proposed food delivery platform according to an embodiment of the present invention.

FIG. 2 shows a flowchart of a method for calculating the system reliability of an online catering distribution system according to an embodiment of the present invention.

FIG. 3 shows a schematic diagram of an apparatus for predicting the reliability of a prediction system according to an embodiment of the present invention.

Herein, the present invention will be described in detail with respect to specific embodiments of the present invention and its viewpoints. Such descriptions are used to explain the structures or steps of the present invention, and are for illustrative purposes rather than limiting the scope of the present invention. Therefore, in addition to the specific embodiments and preferred embodiments in the specification, the present invention can also be widely implemented in other different embodiments. The embodiments of the present invention are described below by specific embodiments, and those skilled in the art can easily understand the efficacy and advantages of the present invention through the contents disclosed in this specification. Moreover, the present invention can also be applied and implemented by other specific embodiments, and various details described in this specification can also be applied based on different requirements, and various modifications or changes can be made without departing from the spirit of the present invention.

The field of the present invention is related to the field of "computing technology". First, a multi-level catering online distribution network is modeled, and a new algorithm including time thresholds and space constraints is proposed to calculate the system for the multi-level catering online distribution network. reliability. System reliability is an important indicator for evaluating the performance of the network, which is defined as the probability that the number of delivery personnel in the system can satisfy the order.

Since the delivery personnel of the catering online distribution system can choose their own work location and time, the multi-level catering online distribution network needs to consider the random characteristics of the delivery personnel in the system, so that the number of delivery personnel in each area presents a multi-state state (multi-state). ), so the system can be constructed as a multi-state network.

The present invention uses a network analysis model to convert the above-mentioned multi-state network considering the random characteristics of delivery personnel, that is, a stochastic online-food delivery system (SODS) into a network topology (network topology). . Then an algorithm is proposed to evaluate the reliability of SODS system.

The present invention focuses on the study of SODS with spatial and temporal constraints, where random capacity (stochastic capacity) represents the number of food delivery personnel available in each region. We evaluate system reliability, which is the probability that SODS will successfully deliver goods to meet each customer's needs within specific space and time constraints.

表示商品的集合。每個節點代表一個區域(例如，一個交付區域)。節點n _i 的當前容量，用x _i 表示，是一個從0到M _i 的整數。容量向量X=(x ₁ ,x ₂ ,…,x _m )。商品用O _a,b ^c,d 表示，索引a表示該商品的發貨節點a，索引b表示該商品的到達節點b，索引c表示該商品的編號，索引d表示該商品是由第d家餐廳製作的。商品的空間權重用W=(w ¹ ,w ² ,...,w ^v )表示，v為商品數量。等待時間用t _d 表示，t _d 是第d家餐廳的準備時間。行駛時間用t _a,b 表示，t _a,b 是從節點i到節點j所消耗的時間。為建構SODS，作出以下假設： First, the Random Catering Online Distribution System (SODS) is composed of nodes, transmission edges (arcs), commodities, commodity's space weight, waiting time, travel time, etc., where each node Represents an area in the delivery area and does not overlap each node; each transit edge (arc) represents a path connecting a pair of nodes. At each node, it may contain several restaurants as well as food delivery personnel servicing the delivery of goods. Items are delivered by one courier, or multiple items can be delivered by a courier. The spatial weight of the product is a ratio given by the restaurant (0≦weight≦1). Let D={ d | d = 1,2,…, r } be a set of restaurants, where d represents the index of each restaurant. The waiting time of the ^dth restaurant is represented by t d , which will vary according to the actual situation of the restaurant. The travel time from node a to node b is obtained by software (such as Google Map), represented by t _{a, b} . The parameters involved in SODS include node capacity, space constraint s , and time constraint T . Each network topology is denoted by G ≡( N , A , D , M , O , W ), representing SODS with outgoing and arriving nodes, where N = { i | i =1,2,…, m } represents the set of nodes, A ={ a _i,j | i,j =1,2,..., m } represents the set of transmission edges connecting node i and node j , M ={ M _i | i =1,2 ,..., m } represents the maximum capacity set of each node,

Represents a collection of items. Each node represents an area (eg, a delivery area). The current capacity of node ni , denoted by _{xi ,} is an integer from 0 to _Mi. The capacity vector X=( x ₁ , x ₂ ,…, x _m ). The product is represented by O _a,b ^c,d , the index a represents the delivery node a of the product, the index b represents the arrival node b of the product, the index c represents the serial number of the product, and the index d represents the product from the dth house. Made by the restaurant. The spatial weight of commodities is represented by W = ( w ¹ , w ² ,..., w ^v ), where v is the number of commodities. The waiting time is denoted by t _d , where t _d is the preparation time for the d -th restaurant. The travel time is denoted by t _a,b , where t _a,b is the time it takes to get from node i to node j . To construct SODS, the following assumptions are made:

1. The capacity of each node is a random variable whose probability is based on a given probability distribution.

2. The travel time t _a,b may be different from t _b,a .

3. The capacity of each node is statistically independent.

4. Each node and transmission edge is perfectly reliable.

Vector operations are performed according to the rules described below:

Y(x ₁ ,x ₂ ,…,x _n )

(y ₁ ,y ₂ ,…,y _n )：x _i

y _i 對於每個i=1,2,…,m。 I.X

Y ( x ₁ , x ₂ ,…, x _n )

( y ₁ , y ₂ ,…, y _n ): x _i

y _i =1,2,…, m for each i .

Y且x _i<y _i ，對於至少一個i。 II. X < Y ( x ₁ , x ₂ ,…, x _n )<( y ₁ , y ₂ ,…, y _n ): X

Y and x _i < y _i for at least one i .

Y既不成立，X<Y也不成立。 III. X is not greater or less than Y : X

Neither Y holds, nor does X<Y.

In the commodity distribution stage, each delivery person must satisfy the constraint condition (1), that is, the commodity's total space weight of each delivery person cannot be greater than the set space constraint s .

Before we distribute the items, we must calculate the minimum number of people to deliver, h . The minimum number of people delivering meals can be obtained according to constraint (2).

。首先，根據商品的空間權重從大到小將這v種商品排列成H*。然後，透過具有約束條件(3)和(4)的所提出的分支定界方法來分配商品。 The branch and bound method is used to analyze all possible cases of the system. After obtaining the minimum number of delivery people, we can allocate the items through constraint (1) and use the proposed branch and bound method to allocate the items. H _f represents the item of the f -th deliveryman, for f = 1, 2,..., h,

. First, arrange the v commodities into H * according to their spatial weights from large to small. Then, the commodities are allocated by the proposed branch and bound method with constraints (3) and (4).

After allocating commodities using the branch and bound method proposed above, the first-stage solutions for commodity allocation can be obtained C=( H _f | f =1,2,… , h ). For each first-level solution C related to commodity allocation, different second-stage solutions related to commodity allocation can be obtained through permutation. R _β =( H _f | f =1,2,…, h ) , which is the βth second-order solution. For each R _β , the transit time of each delivery person in R _β is calculated by constraints (5) and (6). Constraint (5) is used to calculate the transportation time of the delivery person who spends the most time in R _β , and the transportation time of the delivery person is set as the transportation time of R _β . Constraint (6) is used to calculate the transportation time of each delivery person in R _β , and the transportation time is the driving time plus the waiting time. A vector without the first index among the vectors of indices of all dispatching nodes and arriving nodes in R _β is denoted by U _β . The vector that is the index of all stored indices in R _β is denoted by W _β .

S(R _β )=Max(H _f ) for H _f (5) in R _β

S ( H _f )=Σ t _a,b +Σ t ^d for i = U _β , j = L _β , and l = W _β (6)

Then, the present invention proposes two sets F and F' to store feasible and infeasible commodity allocations Hj, respectively. The above two sets are used to quickly compute feasible R.

According to constraints (7) and (8), calculate the capacity vector X = ( x ₁ , x ₂ ,…, x _m ):

X _i =( x ₁ , x ₂ ,..., x _m ), (7)

System reliability R _o is defined as the probability that SODS will successfully deliver v items within space and time constraints. The index l represents the number of capacity vectors. Therefore, the system reliability can be calculated by constraint (9).

Several methods can be used to calculate system reliability, such as state-space decomposition, inclusion-exclusion, disjoint-event methods, and recursive sum of disjoint products disjoint products, RSDP). The RSDP algorithm based on the disjoint product principle (SDP) is more efficient than other methods. So in this paper, we use the RSDP method to calculate the system reliability.

The present invention proposes an algorithm for evaluating the reliability of the SODS system, and the algorithm is described as follows.

Input (Input): network topology G, time constraints, space constraints, waiting time t _d , travel time t _a,b .

Step 1: Calculate the first-level solution C of commodity allocation using the branch and bound method.

1.1) Calculate the minimum number of people to deliver meals according to constraint (10).

1.2) Use H * to obtain the first-level solution C about commodity distribution, where H * is to arrange the v commodities according to their spatial weights from large to small.

Step 2: Use the first-level solution C about commodity distribution to generate a second-level solution R about commodity distribution through permutation.

Step 3: Calculate the transit time of R and delete the infeasible R.

和

。 3.1) Create two sets F and F' . Assume

and

.

3.2) List each courier in R and calculate the shipping time.

3.3) List all feasible R.

Step 4: Use R to obtain the capacity vector. Then remove duplicates and a larger capacity vector.

4.1) Generate a capacity vector X .

4.2) Remove duplicates and larger capacity vectors.

Step 5: Calculate the network reliability using the RSDP method.

Output: R _o

To help understand how the proposed algorithm is implemented, a simple multi-state online food delivery network is shown below. Taking the multi-level state network 100 in the Hsinchu service area of the food delivery platform of FIG. 1 as an example, the present invention can also be used in other practical systems. For the service area can be divided into different delivery areas, the multi-level state network 100 in this case adopts the Activity on Arrow network topology to structure the relative positions of the delivery areas, where a _α represents the divided αth A delivery area, P _{β , ε} represents the shortest direct path from β to ε to the center of the delivery area. The delivery personnel in the delivery platform can choose their own work location, so the number of delivery personnel in each delivery area is considered to be large. stage status. In the present invention, the number of delivery personnel in each delivery area is considered as a multi-level state, and the time threshold and space constraints are considered to evaluate the probability that the current number of delivery personnel on the delivery platform can successfully deliver the order, and this probability is called It is system reliability.

Through historical data, the probability table of delivery personnel capacity for each delivery area can be summarized, as shown in Table 1 below. If the system receives 5 orders, of which the origin and destination are known in advance, the delivery company sets the delivery within the limit of 20 minutes and one delivery box. The relevant order information is shown in Table 2. The delivery area i to The time spent in delivery area j can be known through the delivery system of the delivery company. The details are shown in Table 3.

Therefore, according to the technology of the present invention, the order delivery status that satisfies the space limitation is first obtained through the branch and bound method, then the consumption time is calculated for each order delivery status and the order status that exceeds the time threshold is removed, and then the space and time threshold limitations are passed. The order status is converted to the minimum lower bound vector, and the system reliability is estimated to be 0.9044 through the RSDP method and the minimum lower bound vector. The estimated system reliability indicates that the delivery personnel in the system successfully delivered the delivery within the constraints of 20 minutes and one delivery box. The probability of the following five orders is 0.9044. System reliability can be used as an indicator for a manager to manage the delivery system. Managers can use the system reliability to properly adjust the number of delivery personnel in each delivery area, or increase or decrease the time threshold interval to meet the set service level or The system reliability threshold value can effectively reduce the personnel cost and the risk cost of unsuccessful order delivery through appropriate delivery personnel allocation.

Referring to FIG. 2 , it shows a flowchart of a method for calculating the system reliability of an online catering distribution system according to an embodiment of the present invention. The present invention considers the time threshold and space limitation for the multi-level catering online distribution network, uses the network analysis method to construct a network model, and inputs the time threshold, the space limitation of delivery personnel, the waiting time of the restaurant, the delivery time of each route, etc. as the operation parameters (Step 201), wherein the time threshold can limit the delivery personnel to complete the order delivery at a given time to meet the service level setting of the delivery company, and the space limit can restrict the order quantity of the delivery personnel, so that the goods can be stored and stored in sufficient space. Delivery, restaurant waiting time, and delivery time of each route are related to service indicators and customer satisfaction with the service; then, calculate the minimum number of delivery personnel that satisfies the above space constraints (step 203); Order delivery combination of space constraints, and then find Order status delivered by each courier (step 205); next, calculate the consumption time of each delivery status, and remove the order status exceeding the time threshold (step 207); then, convert the order status into a minimum lower bound vector (step 209) ; Finally, the system reliability is calculated by the recursive sum of disjoint products (RSDP) method (step 211).

Please refer to FIG. 3 , which shows a schematic diagram of an apparatus 300 for predicting system reliability according to an embodiment of the present invention. As shown in the figure, the reliability calculation device of the multi-order state network (MSN) of the present invention may at least include an input device 301 , a memory 303 , a processor 305 and an output device 307 . The input device 301 is electrically connected to the memory 303. The input device 301 may include various input interfaces of a computer device or a file receiving device. The input device 301 can receive nodes related to the Random Catering Online Distribution System (SODS) 3032 through the input device 301. and the structure of the transmission side (ie, the SODS network topology), and store it in the memory 303 . The memory 303 can store the algorithm 3031 of the system reliability calculation method developed for the above-mentioned online food delivery system (SODS). The algorithm 3031 is the same as the process steps disclosed in the foregoing embodiments.

In a preferred embodiment, the memory 303 may include a read-only memory, a flash memory, a disk, or a cloud database.

In a preferred embodiment, the above-mentioned processor 305 is electrically connected to the memory 603. The processor 305 includes a central processing unit, an image processor, a microprocessor, etc., and may include a multi-core processing unit or a plurality of processing units. In combination, the processor 305 can access the SODS network topology 3032 in the memory and the algorithm 3031 of the reliability calculation method to estimate the system reliability.

In a preferred embodiment, the calculation result of the processor 305 can be output by the output device 307 . The output device 307 can be a display for presenting the calculation results, such as an LCD, LED or OLED display screen, or a wired/wireless network transmission device to transmit the calculation results to a remote user.

In short, the present invention takes into account the probability variation characteristics of the number of different delivery personnel in each area per unit time in the practical system, and uses the proposed system reliability as the system. The evaluation index of the system state, the system reliability provided by the present invention can be used to evaluate the state of the existing catering online distribution system of the system, so as to quantify the loadability of the system and strengthen the risk control ability of managers. The number of delivery personnel in each area in the online catering distribution system is a random variable, and does not need to obey a specific probability distribution, so the number of delivery personnel in the online catering distribution system can be easily regarded as a multi-level state. Taking the delivery platform as an example, the number of delivery personnel in the delivery area varies randomly. Under the fluctuation of the order, the probability that the delivery personnel in the system can satisfy the order will also change. Therefore, the reliability of the system proposed in this patent can be achieved through the It is used as a quantitative indicator of system status to actually control the number of delivery personnel to achieve the purpose of successfully delivering all orders. The present invention can provide a performance index of a system related to the distribution industry, the performance index can be used by managers as the basis for dispatching distribution personnel, and when the reliability of the system is insufficient, additional distribution personnel can be dispatched to reduce risk costs.

The protection scope of the present invention is to propose an evaluation method for an online catering distribution system, which can evaluate the reliability of the system under the limitation of space and time thresholds under the condition that the number of distribution personnel is random, which is defined as the given space and time thresholds Under the limitation, the number of delivery personnel in the current system can successfully deliver all orders. According to the assessed system reliability, different numbers of delivery personnel can be dispatched to different regions to reduce personnel costs and the risk of unsuccessful delivery of orders. Therefore, the reliability of the system can provide managers as a management indicator, through which the reliability of the system can be used to dispatch appropriate delivery personnel and to understand the load status of the system. One of the above-mentioned embodiments or any combination of the embodiments can be stored in a computer-readable medium, and can be calculated, processed, calculated, predicted, estimated, etc. through the hardware of the computer, such as a processor, so as to achieve the goal of the present invention. The purpose and effect of desire, and produce unpredictable effects.

The above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention and its benefits are described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that the foregoing embodiments can still be used for Modifications are made to the descriptions, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the claims of the present invention.

201, 203, 205, 207, 209, 211: Steps

Claims (11)

A system reliability evaluation method for an online catering distribution system, the online catering distribution system can be stored in a readable medium, and calculated by a processor, the method comprises the following steps: Through the processor, a network model is constructed by using the network analysis method, and parameters such as time threshold, space limitation of delivery personnel, restaurant waiting time, and delivery time of each route are input; Calculate the minimum number of the delivery personnel that meets the space limitation by the processor, and use the branch-and-bound method to obtain the order delivery combination that meets the space limitation; calculating the elapsed time of each delivery condition by the processor, removing the order condition exceeding the time threshold, and converting the order condition into a minimum lower bound vector; and Calculate the reliability of the system with the processor. The method for evaluating the system reliability of an online catering distribution system according to claim 1, wherein the above-mentioned network model converts the online catering distribution system into a network topology considering the randomness of the above-mentioned delivery personnel. The system reliability evaluation method for an online catering distribution system according to claim 2, wherein the network topology includes a plurality of nodes, a plurality of transmission edges connecting the plurality of nodes, a commodity, a spatial weight of the commodity, the commodity , and the travel time to deliver the item. The system reliability evaluation method for an online catering distribution system according to claim 1, wherein the capacity of each node in the plurality of nodes is a random variable, and its probability is based on a given probability distribution. The method for evaluating the system reliability of an online catering distribution system according to claim 1, wherein the branch-and-bound method is used to find out the order distribution combination that satisfies the space restriction by arranging the products in descending order according to the weights of the products in the order, and then The branch-and-bound method allocates the delivery of the commodities in the order, and after the branch-and-bound method is used to allocate the delivery of the commodities in the order, a first-level solution for commodity allocation is obtained. The method for evaluating the system reliability of an online catering distribution system according to claim 5, wherein each first-level solution of the commodity distribution obtains a different second-level solution for the commodity distribution through substitution. The method for evaluating the system reliability of an online catering distribution system according to claim 1, wherein the above-mentioned calculation of the reliability of the system is evaluated by using a recursive sum of disjoint products (RSDP) algorithm. A system reliability evaluation method for an online catering distribution system, the online catering distribution system can be stored in a readable medium, and calculated by a processor, the method comprises the following steps: Use this processor to build a network model and input parameters such as time threshold, space limit for delivery personnel, restaurant waiting time, and delivery time for each route; using the processor to calculate the minimum number of the delivery personnel that meets the space limit; Use the branch-and-bound method to find the order delivery combination that satisfies the space limitation with the processor, arrange the commodities in descending order according to the weights of the commodities in the order, and then use the branch-and-bound method to allocate the delivery of the commodities in the order, using the above-mentioned branch-and-bound method After allocating the delivery of the commodities in the order by the method, the first-level solution about commodity allocation is obtained; Using the first-level solution for the commodity allocation to generate a second-level solution for the commodity allocation via permutation; The processor calculates the transit time of the second-level solution for the commodity distribution, and deletes the infeasible second-level solution, uses the above-mentioned second-level solution for the commodity distribution to obtain a capacity vector, and then removes duplicates and more. large capacity vectors; and Calculate the reliability of the system with the processor. The method for evaluating the system reliability of an online catering distribution system according to claim 8, wherein the above-mentioned network model converts the online catering distribution system into a network topology considering the randomness of the above-mentioned delivery personnel, and the network topology Pak includes a plurality of nodes, a plurality of transmission edges connecting the plurality of nodes, a commodity, a spatial weight of the commodity, a waiting time of the commodity, and a travel time to transmit the commodity. The method for evaluating the system reliability of an online catering distribution system according to claim 8, wherein calculating the above The shipping time of the second-level solution for the commodity allocation, and the deletion of the second-level solution that is not feasible includes the following steps: Create two sets to store feasible and infeasible allocations of the commodity; List each courier in the second-level solution and calculate the transit time; and List all feasible solutions of this second order. The method for evaluating the system reliability of an online catering distribution system according to claim 8, wherein the above-mentioned calculating the reliability of the system is evaluated by using a recursive sum of disjoint products (RSDP) algorithm.

Images