CN111709570B - Optimization method for network drop-and-drop transport scheduling - Google Patents

Abstract

The invention discloses an optimization method for network drop and pull transportation scheduling, and belongs to the technical field of drop and pull transportation. The optimization method of network drop and pull transportation scheduling is aimed at the situation that the demand at both ends of the operation is unbalanced and there may be multiple drop and pull requirements in drop and pull transportation. For the optimization model of network drop and hang scheduling, a heuristic algorithm is designed to solve the model, and the validity of the model is verified by an example. The present invention fully considers the situation that the demand at both ends is unbalanced and there may be multiple demands in the actual drop and pull transportation, and proposes to carry out network-wide vehicle scheduling optimization in the highway port network, which can be close to the actual application scene and will be beneficial to the drop and pull transportation. Enterprises improve economic efficiency.

Description

An optimization method for network drop and pull transportation scheduling

【Technical field】

The invention relates to the technical field of drop and pull transportation, in particular to an optimization method for network drop and pull transportation scheduling.

【Background technique】

Drop-and-pull transportation, as a way of organizing cargo transportation with both economic and social benefits, has been highly valued by the government and enterprises. Through the pilot project of the project, it is found that drop and pull transportation can significantly increase the actual load rate, improve economic benefits, and reduce average fuel consumption and carbon emissions.

The development of drop and pull transportation needs to rely on the freight depot, which is the starting point and the end point of the return of the drop and pull transportation vehicle. In China, road ports have become the most common freight terminals. The highway port is a comprehensive logistics center in the city. It generally covers a large area and includes services such as drop-and-hang trunk transportation, urban distribution, and temporary storage. In China, Tiandihui highway ports have been deployed in more than 50 cities, with nearly 100 highway ports, and a relatively complete drop and pull transportation network has been formed. Therefore, how to optimize the dispatching of drop and pull vehicles in the whole network has become a practical problem that needs to be solved urgently.

In practical applications, drop and pull transportation is used in many scenarios, including the problem of grocery delivery, garbage collection and transportation, and warehouse delivery of goods to multiple stores. It is generally assumed that the customer has a specific volume of demand, that is, the drop and pull vehicle only needs to visit the customer once to meet the requirements of cargo distribution or garbage collection. However, in the trunk line network of highway port drop and pull transportation, the drop and pull transportation operation is carried out between highway ports, and the highway port is not only the drop and pull operation place, but also the customer demand point. Different from the requirements in the practical application of general drop and pull transportation, the demand for drop and pull at highway ports is based on trailers, and the number of demand may be multiple trailers and the requirements at both ends of the drop and pull may not be equal, that is, there are multiple drop and pull between the two road ports. The situation of unbalanced demand at both ends of drop-and-hang transportation increases the complexity of the optimization of drop-and-hang transportation scheduling.

In order to give full play to its advantages, highway ports need to be connected into lines to form a network. An enterprise relies on the leased or self-built road port site to carry out drop and pull transportation between multiple road ports in many large and medium-sized cities. At this time, the road port is not only a yard for tractors and trailers, but also a drop and pull transportation. demand point. As shown in Figure 1, every two road ports are connected by a channel, and the drop and pull transportation business between multiple road ports has formed a network, but due to the unbalanced demand for drop and pull, and the demand is greater than 1 , the number next to the trailer in Figure 1 indicates the demand for drop and pull transportation in a certain direction. As a result, drop and pull transportation is not a simple one-line two-point two-end or one-line multi-point drop-and-hang scheduling mode, but is dispatched according to demand within the entire network.

Generally speaking, in order to ensure the stability of a certain transportation line, drop-and-hook transportation companies usually arrange some vehicles to specialize in drop-and-hook transportation between two points, and arrange other vehicles to complete other drop-and-hook tasks, so that the network can be transferred to the network. The drop and pull transportation process is divided into two stages.

Stage 1: The two ends of the line are dropped and hung, as shown in Figure 2. Since the organization of the drop-and-hook mode at two points on the line is relatively simple, and the drivers and passengers travel between the two points for a long time, it is beneficial to familiarize themselves with the road conditions and reduce accidents. Therefore, every two points in the whole network can be regarded as independent. Since there is often an unbalanced demand for drop and hang between two points, the smaller demand in the two directions is taken as the amount of work for the drop and hang at both ends of the line and two points. At this time, there must be a demand in a certain direction that has been satisfied. If the need in the other direction is not met, it is moved to the next stage. For example, the numbers next to the trailers in Figure 1 indicate the drop and pull demand, that is, 5 trailers need to be dropped and pulled from point 1 to point 2, and 3 trailers need to be dropped and pulled from point 2 to point 1. At this time, the demand between points 1 and 2 is different. Balanced, take the smaller amount of demand between two points 3 as the amount of drop and hook work at both ends of the line and two points, then the drop and pull demand from point 2 to point 1 is satisfied, and the remaining needs from point 1 to point 2 still have 2 trailers If it is not satisfied, it needs to be scheduled in the whole network and transferred to the next stage.

Stage 2: The whole network loops and hangs, as shown in Figure 3. After the drop and hang scheduling in phase 1, there is still a phenomenon that the unidirectional demand is not met between the two points, and the scheduling arrangement needs to be carried out in the whole network. As shown in Figure 3, there are 2, 1, 1, 2, and 2 trailers between points 1 to 2, 1 to 4, 1 to 5, 2 to 3, and 2 to 4, respectively. Shipping is required. Due to the limited operating time of the tractor every day, it is necessary to make a scheduling plan that minimizes the cost of network scheduling and the purchase of tractors at this stage.

Therefore, in view of the characteristics of drop-and-haul transportation in the highway port network, the present invention establishes a drop-and-hang scheduling optimization model for the entire network, considering the complex situation that customers have multiple drop-and-hang requirements and the unbalanced demand at both ends of the drop-and-hang in actual drop-and-haul transportation. Provide a reference for the actual highway port network drop and pull optimization (efficiency improvement).

[Content of the invention]

In view of the above problems, the present invention provides an optimization method for network drop and pull transportation scheduling. The optimization method fully considers the situation that the demand at both ends is unbalanced and the demand may be multiple in actual drop and pull transportation. The network-wide vehicle scheduling optimization can be close to the actual application scenario, which will help drop and pull transportation enterprises to improve economic benefits.

For achieving the above object, the technical scheme adopted in the present invention is as follows:

An optimization method for network drop and pull transport scheduling, which is applied to a road port network drop and pull transport composed of tractors, trailers and road ports, and includes the following steps:

(1) Establish an optimization model for network drop-and-hang scheduling objectives;

(2) Use a two-stage heuristic algorithm to solve the objective optimization model described in step (1):

First, the drop and pull demand q _ij between any two highway ports is divided into two parts q' _ij and q"_ij; if q _ij ≥ q _ji , then q' _ij =q _ji =q' _ji , q" _ij = q _ij -q _ji , q" _ji =0; if q _ij <q _ji , then q' _ij =q _ij =q' _ji , q" _ij =0, q" _ji =q _ji -q _ij , q' _ij The formed matrix is a symmetric matrix; among them, q _ij : represents the drop and pull transport demand of road ports i to j, expressed by the number of times of drop and pull transport; q _ji : represents the drop and pull transport demand of road ports j to i, expressed by the drop and pull transport times It is represented by the number of transport trips; since the demand is not necessarily balanced, q _ij is not necessarily equal to q _ji ;

Stage 1: Scheduling the demand q' _ij drop and pull demand: adopt the scheduling of drop and pull vehicles at both ends of the line, that is, select a highway port in the network arbitrarily, and start the task assignment of tractor drop and pull, and the tractor is in the two highway ports Flip back and forth between heavy loads;

Stage 2: Scheduling the q” _ij drop and pull demand after the first stage scheduling: adopt the whole network loop drop and pull vehicle scheduling, that is, select the line with the least remaining drop and pull demand after the first stage of scheduling to first perform the drop and pull of this stage. Hang up, and take the singular demand starting point as the starting point of the tractor.

The present invention also provides another optimization method for network drop and pull transport scheduling, which is applied to the drop and pull transport of the road port network composed of tractors, trailers and road ports, and includes the following steps:

(1) Establish an optimization model for network drop-and-hang scheduling objectives;

(2) Use the heuristic algorithm to solve the objective optimization model described in step (1): use the pure network loop drop and pull to perform vehicle scheduling on the drop and pull demand q _ij between any two highway ports.

Preferably, the network drop-and-hang scheduling objective optimization models described in step (1) of the above two methods are:

Objective function:

空驶成本为：

当天使用车辆固定成本为Z_固＝Kc；Assume that the network operation cost is composed of the cost of re-hook driving, empty driving cost and fixed cost, and the cost of re-hook driving

The cost of empty driving is:

The fixed cost of using the vehicle on that day is Z _solid = Kc;

Then the goal is to minimize the network operation cost, which is expressed as:

The constraints are:

Drop and hook requirements are met:

Continuous working time limit of tractor:

In the formula, i, j, l: represents the number of the road port, i, j, l∈N;

K: Represents the set of tractors, K={1,2,...,k}, and K also represents the total number of tractors required by the road network;

k: indicates the number of the tractor, k∈K;

d _ij : represents the distance between highway ports i and j, unit: km;

q _ij : Indicates the drop and pull transport demand from road ports i to j, expressed by the number of drop and pull transport trips;

C _ij , c _ij : respectively represent the cost of tractor towing heavy-hook and empty driving, unit: yuan/km;

c: Indicates the fixed cost of the tractor used per unit time, including vehicle depreciation and employee wages, unit: yuan;

T: Indicates the continuous working time of the tractor, which stipulates that all tractors work continuously for the same time every day;

v ₁ , v ₂ : respectively indicate the speed of the tractor towing the heavy-hook and empty driving, unit: km/h;

k ₁ , k ₂ : respectively represent the state of the tractor towing heavy suspension and empty driving, k∈K;

分别表示牵引车从公路港重载和空驶驶往公路港的趟数。

Respectively represent the number of trips of the tractor from the road port to the road port with heavy load and empty.

Preferably, the stage 1 described in step (2) is carried out as follows:

The first step: initial parameter setting, let i=1, K=0, Z _weight =0, Z _empty =0, Z _solid =0;

那么K＝K+1，Z_固＝Kc，q'_ij＝0＝q'_ji，且若q”_ij＞0，则仍然可以选择公路港i，j为间任务进行运输，若还有工作时间剩余，可以选取公路港j出发距离次小(d_jl)的l作为下一个甩挂点，直到工作时间用完为止，其中(d_jl)表示公路港j至公路港l的距离，q”_ij＝q”_ij-1，Z_重＝Z_重+2q'_ijd_ijC_ij+d_ijC_ij+d_jlC_jl+...；若q”_ij＝0，则选取公路港i出发距离次小(d_il)的l作为下一个甩挂点，直到车辆工作时间用完为止，对应的q'_jl＝q'_jl-1，Z_重＝Z_重+2q'_ijd_ijC_ij+d_ilC_il+...；如果q'_ij＞0且d_ij＜v₁T，且如果

则公路港i，j之间还需要安排其他车辆进行运输，需要车辆数为

有剩余能力的车辆仍然可以完成其它甩挂任务，那么q'_ij＝0＝q'_ji，

Z_重＝Z_重+2q'_ijd_ijC_ij，Z_固＝Kc；The second step: select q' _ij > 0, and the highway port i, j corresponding to max(d _ij ) is the end point of the first vehicle service; if q' _ij > 0 and d _ij ≥ v ₁ T, Then K=K+2q' _ij , q' _ij =0=q' _ji , Z _weight =Z _weight + 2q' _ij d _ij C _ij , Z _solid =Kc; if q' _ij >0 and d _ij <v ₁ T, and if

Then K=K+1, Z = _Kc , q' _ij = 0 = q' _ji , and if q" _ij > 0, you can still choose road port i, j as the task for transportation, if there is still working time For the rest, the l with the next smallest departure distance (d _jl ) from the highway port j can be selected as the next drop and hang point until the working time is used up, where (d _jl ) represents the distance from the highway port j to the highway port l, q” _ij =q" _ij -1, Z _weight = Z _weight + _2q ' _ij d _ij C _ij +d _ij C _ij +d _jl C _jl +...; The l with a small (d _il ) is used as the next drop point until the working time of the vehicle runs out, the corresponding q' _jl =q' _jl -1, Z _weight = Z _weight + 2q' _ij d _ij C _ij +d _il C _il +...; if q' _ij > 0 and d _ij < v ₁ T, and if

Then other vehicles need to be arranged for transportation between highway ports i and j, and the number of vehicles required is

Vehicles with remaining capacity can still complete other drop-and-hang tasks, then q' _ij =0=q' _ji ,

Z _weight =Z _weight +2q' _ij d _ij C _ij , Z _solid =Kc;

Step 3: For all i∈N, i=i+1, if q' _ij > 0, go back to the second step to calculate; if all q' _ij = 0, stop the calculation, Z ₁ '=Z _re + Z _solid .

Preferably, stage 2 described in the two-stage heuristic algorithm of step (2) is performed as follows:

The first step: choose i∈N;

The second step: select the highway port i with q” _ij > 0, j is the demand point for drop and pull dispatch; if d _ij ≥ v ₁ T, then K=K+q” _ij , q” _ij =0, Z _weight = Z _Re +q” _ij d _ij C _ij , Z = _Kc ; if d _ij <v ₁ T, there are two strategies:

若还有剩余能力的车辆仍然可以完成其它甩挂任务，则

q”_ij＝0，Z_重＝Z_重+q”_ijd_ijC_ij，Z_空＝Z_空+(q”_ij-1)d_jic_ji，Z_固＝Kc；Strategy 1 is that vehicle k travels to and from road ports i and j. At this time, there is only one-way loading of goods, and road ports j to i are empty; if vehicle k can complete q” _ij drop and hang within time T, then K=K+ 1, q" _ij = 0, Z _heavy = Z _heavy + q" _ij d _ij C _ij , Z _empty = Z _empty + (q" _ij -1)d _ji c _ji , Z _solid = Kc, if there is still working time For the remainder, you can select port l where q” _jl > 0 starts from road port j as the next drop-and-hang point; if vehicle k cannot complete the drop and drop of q” _ij within time T, other vehicles need to be dispatched to perform the drop-and-hang. The number of vehicles required to complete the drop-and-hang task between highway ports i and j is

If vehicles with remaining capacity can still complete other drop-and-hang tasks, then

q" _ij = 0, Z _heavy = Z _heavy + q" _ij d _ij C _ij , Z _empty = Z _empty + (q " _ij -1) d _ji c _ji , Z _solid = Kc;

Strategy 2 is that vehicle k completes the drop and pull task in the highway port network. At this time, there are heavy loads and empty loads in the entire network. Vehicle k completes a drop and pull transport from highway port i to j, q” _ij = q” _ij -1, and then select the port l where q” _jl > 0 starts from the highway port j as the next drop and hang point, q” _jl = q” _jl -1, until there is no remaining working time, K = K + 1, Z _weight = Z _weight +d _ij C _ij +d _jl C _jl +...; if starting from road port j, all q" _jl = 0, then empty drive to port l, and then select the road starting from port l with drop-and-hang task The port will drop and hang until there is no remaining working time, K=K+1, Z _weight =Z _weight +d _ij C _ij +..., Z _empty = Z _empty + d _jl c _jl +..., Z _solid =Kc;

Step 3: For all i∈N, i=i+1, if q” _ij > 0, go back to the second step to calculate; if all q” _ij = 0, stop the calculation, Z ₁ ”=Z _re + Z _empty + Z _solid ;

Therefore, the total cost of network drop and hang scheduling is: Z ₁ =Z' ₁ +Z ₁ ”.

Preferably, in step (2), the pure network loop drop and pull is used to perform vehicle scheduling on the drop and pull demand q _ij between any two highway ports according to the following steps:

The first step: choose i∈N;

The second step: select the highway port i and j with q _ij >0 as the demand point for drop and pull dispatch; if d _ij ≥ v ₁ T, then K=K+q _ij , q _ij =0, Z _weight =Z _weight + q _ij d _ij C _ij , Z = _Kc ; if d _ij <v ₁ T, vehicle k completes the drop-and-hang task in the highway port network, at this time there is heavy load and no load in the whole network, vehicle k completes highway port i One drop and pull transport to j, q _ij = q _ij -1, and then select port l where q _jl > 0 from road port j as the next drop and pull point, q _jl = q _jl -1, until there is no remaining working time So far, K=K+1, Z _weight =Z _weight +d _ij C _ij +d _jl C _jl +...; if all q _jl = 0 starting from road port j, then you have to empty to port l, and then choose from The highway port with the drop-and-hang task departing from the l port will be dropped and hung until there is no remaining working time, K=K+1, Z _weight =Z _weight +d _ij C _ij +..., Z _empty = Z _empty + d _jl c _jl +..., Z _solid = Kc;

Step 3: For all i∈N, i=i+1, if q _ij >0, go back to the second step for calculation; if all q _ij =0, stop the calculation, Z ₁ =Z _heavy +Z _empty + Z _solid .

Therefore, the total cost of pure network drop and hang scheduling is: Z ₁ = Z _heavy + Z _empty + Z _solid .

By adopting the above-mentioned technical scheme, the beneficial effects of the present invention are:

The present invention establishes a whole network drop and pull scheduling optimization model, designs a heuristic algorithm to solve the model, and verifies the validity of the model with an example, so that the drop and pull transportation optimization in the network can reduce the number of traction vehicles used and reduce the overall transportation. cost. And in the network drop and pull transportation, scheduling optimization within the pure network range can more effectively reduce the number of traction vehicles used and reduce transportation costs. The present invention fully considers the situation that the demand at both ends is unbalanced and there may be multiple demands in the actual drop and pull transportation, and proposes to carry out network-wide vehicle scheduling optimization in the highway port network, which can be close to the actual application scene and will be beneficial to the drop and pull transportation. Enterprises improve economic efficiency.

【Description of drawings】

Fig. 1 is the schematic diagram of drop and pull transportation of highway port network in the prior art;

Fig. 2 is a schematic diagram of the two-point drop-and-hang on both ends of a line in stage 1 in the prior art;

Fig. 3 is the schematic diagram of the whole network cycle drop and hang of stage 2 in the prior art;

【Detailed ways】

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The invention mainly includes the steps of assuming a model, establishing a model, and using a two-stage heuristic algorithm to solve the model.

First, the model assumptions:

According to the characteristics of drop and pull transport in the highway port network, the following assumptions are made:

(1) All tractors and trailers in the network are standard models, and the models are the same;

(2) There are enough trailers in each road port in the network, this assumption can be achieved by the integration of social resources of the road port drop and pull transportation enterprise;

(3) The operation time of the tractor's drop-trailer trailer is not considered, and the number of trailers is sufficient, and there is no need for the tractor to wait for the trailer loading and unloading operations;

(4) The distance and time for round-trip travel between the two highway ports are the same;

(5) The demand for drop and hang is generated every day in the network, and the demand still fluctuates every day. The enterprise can arrange scheduling according to the actual demand every day. Due to the long distance between some nodes, the drop-and-hang transportation may not be completed within one day, so it is assumed that the drop-and-hang task that was not completed the previous day will not affect the subsequent scheduling.

Second, build the model:

According to the actual operation of the highway port network, the drop-and-trail transportation enterprise pursues the minimum operating cost of the entire network, and establishes a model with the goal of meeting the demand while investing the minimum number of tractors.

空驶成本为：

当天使用车辆固定成本为：Z_固＝Kc；目标为网络运营成本最小表示为：(1) Goal 1: The network operation cost is composed of the cost of re-hook driving, empty driving cost and fixed cost. The cost of re-hook driving is

The cost of empty driving is:

The _fixed cost of the vehicle used on the day is: Z = Kc; the target is the minimum network operating cost expressed as:

(2) Goal 2: The minimum number of tractors put into use is expressed as:

min Z ₂ =K

(3) Constraints:

Drop and hook requirements are met:

Continuous working time limit of tractor:

如果每一个甩挂需求都安排一辆牵引车进行牵引，显然

即目标2可以变成目标1的一个约束条件。The network drop and hang scheduling problem with unbalanced demand is a dual-objective problem. Goal 1 seeks to minimize the network operation cost. Since the drop-and-hang requirements must be met, under this goal, the main trade-off between the empty cost of the tractor and the fixed cost is to be weighed. The fixed cost Kc of the tractor is proportional to the number of tractors invested K. If the tractor idle cost is greater than the fixed cost, it will increase the tractor investment and reduce the overall operating cost, which contradicts goal 2; if the tractor idle cost is less than the fixed cost cost, then the tractor will be arranged to run empty without increasing the number of tractors invested, which is consistent with goal 2. At this time, only a non-inferior solution that satisfies objective 1 and objective 2 can only be found, and the drop-and-hang requirement of the whole network is

If a tractor is arranged for every drop and pull request, obviously

That is, goal 2 can become a constraint of goal 1.

From the above, the final change of the network drop and hang scheduling model with unbalanced demand is:

Constraints ST:

Among them, the symbol description is:

i, j, l: indicates the number of the road port, i, j, l∈N;

K: Represents the set of tractors, K={1,2,...,k}, and K also represents the total number of tractors required by the road network;

k: indicates the number of the tractor, k∈K;

d _ij : represents the distance between highway ports i and j, unit: km;

q _ij : Indicates the drop and pull transport demand from road ports i to j, expressed by the number of drop and pull transport trips;

C _ij , c _ij : respectively represent the cost of towing heavy-hook and empty driving of tractors in highway ports i to j, unit: yuan/km;

c: Indicates the fixed cost of the tractor used per unit time, including vehicle depreciation and employee wages, unit: yuan;

T: Indicates the continuous working time of the tractor, which stipulates that all tractors work continuously for the same time every day;

v ₁ , v ₂ : respectively indicate the speed of the tractor towing the heavy-hook and empty driving, unit: km/h;

k ₁ , k ₂ : respectively represent the state of the tractor towing heavy suspension and empty driving, k∈K;

分别表示牵引车从公路港重载和空驶驶往公路港的趟数。

Respectively represent the number of trips of the tractor from the road port to the road port with heavy load and empty.

Finally, a two-stage heuristic algorithm is used to solve the model, including algorithm ideas and specific steps of the algorithm.

Algorithm idea:

The drop-and-pull scheduling problem of the highway port network with unbalanced demand is essentially a vehicle scheduling problem with multiple depots. However, since each highway port can be regarded as both a depot and a customer, it is also a special dispatch in which the depot and the demand point overlap. question. Since the vehicle cannot determine the starting point and the ending point, starting from any highway port, there are multiple paths to choose from, which also increases the difficulty of solving. Scheduling can be carried out in the whole network, or according to the conventional practice, the drop and drop demand q _ij between any two highway ports can be divided into two parts q' _ij and q" _ij (separation is not considered in pure network drop and drop) .If q _ij ≥q _ji , then q' _ij =q _ji =q' _ji ,q" _ij =q _ij -q _ji ,q" _ji =0; if q _ij <q _ji , then q' _ij =q _ij =q' _ji , q" _ij =0, q" _ji =q _ji -q _ij , the matrix formed by q' _ij is a symmetric matrix. For the requirements q' _ij and q" _ij , a two-stage heuristic is correspondingly designed Algorithm needle to solve.

(

表示向上取整)，如果a'_ijk＞2q'_ij＝q'_ij+q'_ji，则第k辆能完成公路港i和j之间的所有任务，还可以选择i至其它有甩挂需求且距离次大的点再进行甩挂。如果a'_ijk≤2q'_ij，则第k辆不能完成公路港i和j之间的所有任务，还需要选派其它车辆来共同完成，完成公路港i和j之间的甩挂需求需要派出

辆牵引车，有剩余能力的车辆仍然可以完成其它甩挂任务。Stage 1: Dispatch of drop-and-trail vehicles at both ends of the line. In this stage, only q' _ij needs to be scheduled. Randomly select a road port in the network, and start the task assignment of the tractor to drop and hang. The tractor can be fully utilized between the two road ports with heavy loads and back and forth. If the distance d _ij between highway ports i and j is large, the tractor can only complete one trip during the working time T of one day (according to the assumption, even if one trip is not completed, one trip will be If the distance d _ij between highway ports i and _j is small, the k-th tractor can complete the work within one day’s working time T If there are multiple trips of transportation, then the working time can be fully utilized, and the number of towing trips that can be completed within T time is:

(

Represents rounded up), if a' _ijk >2q' _ij =q' _ij +q' _ji , then the kth vehicle can complete all tasks between highway ports i and j, and you can also choose i to other ones that have a drop-and-hang requirement And the distance is the next largest point and then hang up. If a' _ijk ≤ 2q' _ij , then the k-th vehicle cannot complete all tasks between highway ports i and j, and other vehicles need to be selected to complete them together.

a tractor, vehicles with remaining capacity can still complete other drop-and-hook tasks.

则牵引车k还可以继续牵引下一条路径；若

则路径j至l不能由k来甩挂，需要选择j至其它的有甩挂需求且距离次大的点再进行判断。Stage 2: Full network loop drop and pull vehicle scheduling (if the drop and drop demand q _ij is not divided into two parts q' _ij and q" _ij , that is, the drop and drop of the demand q _ij is directly considered, a pure network can be designed according to the idea of this stage Drop-and-trail vehicle scheduling algorithm). At this stage, q" _ij is scheduled, and there may be a situation where the tractor is empty. Because in the tractor scheduling, the ideal state is that after the tractor completes the heavy-load drop and pull of a line, it can continue to carry out the heavy-load drop and drop of the next line. Therefore, select the line with the least remaining drop and pull demand to perform the drop and pull of this stage first. and take the singular demand starting point i as the starting point of the tractor. If the distance d _ij between highway ports i and j is relatively large, and the tractor can only complete one trip during the working time T of a day, then highway ports i to j need q” _ij tractors for towing; If the distance d _ij between highway ports i and j is small, multiple towing trips can be completed within T _k time, since there is only the drop and pull demand q” _ij from i to j between i and j at this stage, And q" _ji = 0, so after the tractor k completes one drop and pull from i to j, there is no need for return drop and drop, but needs to find the next drop and drop path, starting from road port j to select j to other drop and pull paths demand and the point l with the largest distance, if

Then the tractor k can continue to tow the next path; if

Then the path j to l cannot be dropped and hung by k, and it is necessary to select j to other points with the second largest distance and need to be dropped and then judged.

The specific steps of the algorithm:

The drop-and-hang requirement q _ij is divided into two parts q' _ij and q" _ij .

Stage 1: Scheduling the demand q' _ij drop and hang demand.

The first step: initial parameter setting, let i=1, K=0, Z _weight =0, Z _empty =0, Z _solid =0.

那么K＝K+1，Z_固＝Kc，q'_ij＝0＝q'_ji，且若q”_ij＞0，则仍然可以选择公路港i，j为间任务进行运输，若还有工作时间剩余，可以选取公路港j出发距离次小(d_jl)的l作为下一个甩挂点，直到工作时间用完为止，其中(d_jl)表示公路港j至公路港l的距离，q”_ij＝q”_ij-1，Z_重＝Z_重+2q'_ijd_ijC_ij+d_ijC_ij+d_jlC_jl+...；若q”_ij＝0，则选取公路港i出发距离次小(d_il)的l作为下一个甩挂点，直到车辆工作时间用完为止，其中(d_il)表示公路港i至公路港l的距离，对应的q'_jl＝q'_jl-1，Z_重＝Z_重+2q'_ijd_ijC_ij+d_ilC_il+...。如果q'_ij＞0且d_ij＜v₁T，且如果

则公路港i，j之间还需要安排其他车辆进行运输，需要车辆数为

有剩余能力的车辆仍然可以完成其它甩挂任务，那么q'_ij＝0＝q'_ji，

Z_重＝Z_重+2q'_ijd_ijC_ij，Z_固＝Kc。The second step: select q' _ij > 0, and the highway port i, j corresponding to max(d _ij ) is the end point of the first vehicle service. If q' _ij >0 and d _ij ≥ v ₁ T, then K=K+2q' _ij , q' _ij =0=q' _ji , Z _weight =Z _weight +2q' _ij d _ij C _ij , Z _solid = Kc. if q' _ij > 0 and d _ij < v ₁ T, and if

Then K=K+1, Z = _Kc , q' _ij = 0 = q' _ji , and if q" _ij > 0, you can still choose road port i, j as the task for transportation, if there is still working time For the rest, the l with the next smallest departure distance (d _jl ) from the highway port j can be selected as the next drop and hang point until the working time is used up, where (d _jl ) represents the distance from the highway port j to the highway port l, q” _ij =q" _ij -1, Z _weight = Z _weight + _2q ' _ij d _ij C _ij +d _ij C _ij +d _jl C _jl +...; The l with a small (d _il ) is used as the next drop point until the working time of the vehicle runs out, where (d _il ) represents the distance from the highway port i to the highway port l, and the corresponding q' _jl = q' _jl -1, Z _weight = Z _weight + 2q' _ij d _ij C _ij + d _il C _il + . . . if q' _ij > 0 and d _ij < v ₁ T, and if

Then other vehicles need to be arranged for transportation between highway ports i and j, and the number of vehicles required is

Vehicles with remaining capacity can still complete other drop-and-hang tasks, then q' _ij =0=q' _ji ,

Z _weight =Z _weight +2q' _ij d _ij C _ij , Z _solid =Kc.

Step 3: For all i∈N, i=i+1, if q' _ij > 0, go back to the second step to calculate; if all q' _ij = 0, stop the calculation, Z ₁ '=Z _re + Z _solid .

Stage 2: Scheduling the q" _ij drop-and-hang requirements after the first stage scheduling.

The first step: choose i∈N;

若还有剩余能力的车辆仍然可以完成其它甩挂任务，则

q”_ij＝0，Z_重＝Z_重+q”_ijd_ijC_ij，Z_空＝Z_空+(q”_ij-1)d_jic_ji，Z_固＝Kc。策略2是车辆k在公路港网络中完成甩挂任务，此时整个网络中存在重载也存在空载，车辆k完成公路港i至j的一趟甩挂运输，q”_ij＝q”_ij-1，再选取公路港j出发q”_jl＞0的l港作为下一个甩挂点，q”_jl＝q”_jl-1，直到工作时间没有剩余为止，K＝K+1，Z_重＝Z_重+d_ijC_ij+d_jlC_jl+···。若从公路港j出发所有q”_jl＝0，那么要空驶至l港，再选取从l港出发的有甩挂任务的公路港进行甩挂，直到工作时间没有剩余为止，K＝K+1，Z_重＝Z_重+d_ijC_ij+···，Z_空＝Z_空+d_jlc_jl+···。The second step: select the highway port i with q” _ij > 0, j is the demand point for drop and pull dispatch. If d _ij ≥ v ₁ T, then K=K+q” _ij , q” _ij =0, Z _weight = Z _Weight +q” _ij d _ij C _ij , Z _solid =Kc. If d _ij < v ₁ T, there are two strategies at this time. Strategy 1 is that vehicle k travels back and forth between road ports i and j. At this time, there is only one-way loading of goods, and road ports j to i are empty. If vehicle k can complete q” _ij drop and hang amount within time T, then K=K+1, q” _ij =0 Z _heavy =Z _heavy +q” _ij d _ij C _ij , Z _empty =Z _empty +(q” _ij -1) d _ji c _ji , Z _solid = Kc, if there is still working time left, you can select the port l where the road port j starts from q" _jl > 0 as the next drop and pull point. If the vehicle k cannot be within the time T To complete the drop and pull amount of q” _ij , other vehicles need to be sent to drop and pull. The number of vehicles required to complete the drop and pull task between highway ports i and j is

If vehicles with remaining capacity can still complete other drop-and-hang tasks, then

q" _ij = 0, Z _heavy = Z _heavy + q" _ij d _ij C _ij , Z _empty = Z _empty + (q" _ij -1) d _ji c _ji , Z _solid = Kc. Strategy 2 is that vehicle k is on the road The drop and pull task is completed in the port network. At this time, there are heavy loads and empty loads in the entire network. Vehicle k completes a drop and pull transport from road port i to j, q” _ij = q” _ij -1, and then selects the road port. j starts from port l where q” _jl > 0 as the next drop and hang point, q” _jl = q” _jl -1, until there is no remaining working time, K = K + 1, Z _weight = Z _weight + d _ij C _ij +d _jl C _jl +.... If starting from road port j, all q” _jl = 0, then empty sailing to port l, and then select the road port starting from port l with drop-and-hang task for drop-and-hang until there is no remaining working time, K=K+1 , Z _weight = Z _weight + d _ij C _ij + ···, Z _empty = Z _empty + d _jl c _jl + ···.

Step 3: For all i∈N, i=i+1, if q” _ij > 0, go back to the second step to calculate; if all q” _ij = 0, stop the calculation, Z ₁ ”=Z _re + Z _empty + Z _solid .

Therefore, the total cost of network drop and hang scheduling is: Z ₁ =Z' ₁ +Z ₁ ”.

The pure network loop drop and pull is used to dispatch vehicles to the drop and pull demand q _ij between any two highway ports as follows:

The first step: choose i∈N;

The second step: select the highway port i and j with q _ij >0 as the demand point for drop and pull dispatch; if d _ij ≥ v ₁ T, then K=K+q _ij , q _ij =0, Z _weight =Z _weight + q _ij d _ij C _ij , Z = _Kc ; if d _ij <v ₁ T, vehicle k completes the drop-and-hang task in the highway port network, at this time there is heavy load and no load in the whole network, vehicle k completes highway port i One drop and pull transport to j, q _ij = q _ij -1, and then select port l where q _jl > 0 from road port j as the next drop and pull point, q _jl = q _jl -1, until there is no remaining working time So far, K=K+1, Z _weight =Z _weight +d _ij C _ij +d _jl C _jl +...; if all q _jl = 0 starting from road port j, then you have to empty to port l, and then choose from The highway port with the drop-and-hang task departing from the l port will be dropped and hung until there is no remaining working time, K=K+1, Z _weight =Z _weight +d _ij C _ij +..., Z _empty = Z _empty + d _jl c _jl +..., Z _solid = Kc;

Step 3: For all i∈N, i=i+1, if q _ij >0, go back to the second step for calculation; if all q _ij =0, stop the calculation, Z ₁ =Z _heavy +Z _empty + Z _solid .

Therefore, the total cost of pure network drop and hang scheduling is: Z ₁ = Z _heavy + Z _empty + Z _solid .

In this application, the parameters with the subscript jl represent the parameters from road port j to road port l, for example (d _jl ) represents the distance from road port j to road port l, and q _jl represents the distance from road port j to road port l drop-off demand. C _jl and c _jl respectively represent the cost of tractor towing heavy hitch and empty driving from road port j to road port l. Since the equivalent letters (parameters) have the same meaning, and the different subscripts only represent different highway ports, this is a common knowledge in the field, so it will not be repeated here. The same is true for parameters with subscripts il.

Example

Relying on Tiandihui Highway Port, logistics enterprise L has drop and pull transportation points in 10 cities. How to arrange vehicle scheduling in the drop and pull network to minimize the cost of the entire network is a technical problem that enterprise managers hope to solve.

Assuming that the cost of heavy-hung and no-load running of the tractor is C _ij = 0.8 yuan/km, , c _ij = 0.4 yuan/km, the speed of the tractor's heavy-hook and no-load driving is v ₁ = 50km/h, v ₂ = 70km/h , the fixed cost per day is c=500 yuan, the continuous working time per day is T=16 hours, and the value of d _ij is obtained according to the Google map, as shown in Table 1. The q _ij value uses the randint function in the Matlab 2010a software to generate drop-and-hang transportation tasks between 10 customer points in [0,10]. All drop-and-hang transportation tasks are integers, as shown in Table 2.

According to the above two-stage heuristic algorithm, MATLAB 2010a is used for programming calculation, and the results are shown in Table 3. Among them, the traditional method refers to sending a truck for each transportation task (generally not used, here is for comparison); pure network transportation refers to assigning transportation tasks in the entire network without dividing into two stages. As long as the vehicle has the remaining transportation time, it can carry out the transportation of the next task, which can be solved according to the idea of strategy 2 in stage 2 of the heuristic algorithm.

Table 1 Distance table between highway ports (unit, km)

serial number A B C D E F G H I J A 0 1218 680 120 290 1080 600 690 945 1010 B 1218 0 540 1200 1500 2300 1710 1800 2000 2120 C 680 540 0 660 960 1760 1170 1260 1170 1590 D 120 1200 660 0 314 1108 525 609 818 930 E 290 1500 960 314 0 816 402 586 846 964 F 1080 2300 1760 1108 816 0 661 832 1131 1128 G 600 1710 1170 525 402 661 0 274 594 600 H 690 1800 1260 609 586 832 274 0 323 356 I 945 2000 1170 818 846 1131 594 323 0 276 J 1010 2120 1590 930 964 1128 600 356 276 0

Table 2 Drop and drop demand table between highway and port (unit: vehicle)

serial number A B C D E F G H I J A 0 7 8 0 1 4 6 2 0 1 B 9 0 2 0 10 0 3 3 2 7 C 1 3 0 5 0 9 8 5 8 5 D 10 0 7 0 8 10 2 2 0 8 E 6 1 9 10 0 5 7 9 10 7 F 1 9 10 1 9 0 2 2 8 9 G 3 7 6 6 0 3 0 2 5 9 H 6 3 1 5 4 9 6 0 6 3 I 10 10 1 0 2 4 8 2 0 7 J 10 0 2 3 8 1 0 4 5 0

Table 3 Comparison of calculation results

It can be seen from Table 3 that both the two-stage network drop-and-hook and pure network drop-and-hook methods use fewer tractors and lower costs than the traditional method. The two-stage network drop and pull strategy 1, strategy 2, and pure network drop and pull reduce the number of tractors by 7%, 10.4%, and 27.1%, respectively, and save 0.8%, 0.7%, and 8.6% of the cost. Compared with the traditional method, in the two-stage network drop and pull and pure network drop and pull, the tractor is allowed to tow multiple trailer tasks if possible, so the number of tractors used is bound to be reduced. And when the increased cost of empty tractor driving is less than the fixed cost saved due to the reduction in the number of tractors, the cost of both drop-and-trail networks shows a downward trend.

Compared with the two-stage network drop and pull, the pure network drop and pull transportation has better performance in terms of the number of tractors and the cost. Compared with strategy 1 and strategy 2 of the two-stage network drop and pull, the number of pure network drop and pull tractors is reduced by 21.6% and 18.6%, and the cost is saved by 7.8% and 7.9%. It shows that scheduling optimization in the whole network can effectively reduce the purchase of tractor vehicles and reduce the overall cost. That is to say, operators of drop-and-hang enterprises should change the original mode of network drop-and-hang in two stages, and adopt whole-network scheduling optimization to achieve better operation performance, thereby reducing costs and increasing profits.

Comparing the two strategies of two-stage network drop and hang, because the stage 2 of strategy 2 is to perform scheduling optimization in the whole network, while the stage 2 of strategy 1 is to optimize between two ends before turning to network optimization, which also verifies the whole network. Vehicle scheduling can reduce the number of tractors. Although the number of tractors in strategy 2 is slightly less than that in strategy 1, strategy 1 is slightly better than strategy 2 in cost performance because strategy 2 has a 44% increase in no-load cost compared to strategy 1.

To sum up, compared with traditional transportation methods, optimizing drop and pull transportation in the network can reduce the number of traction vehicles used and reduce the overall transportation cost. And in the network drop and pull transportation, scheduling optimization in the entire network can more effectively reduce the number of traction vehicles used and reduce transportation costs. The present invention fully considers the situation that the demand at both ends is unbalanced and there may be multiple demands in the actual drop and pull transportation, and proposes to carry out network-wide vehicle scheduling optimization in the highway port network, which can be close to the actual application scene and will be beneficial to the drop and pull transportation. Enterprises improve economic efficiency.

The above description is a detailed description of the preferred feasible embodiments of the present invention, but the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications completed under the technical spirit suggested by the present invention shall belong to This invention covers the scope of the patent.

Claims (5)

1. A network drop and pull transport scheduling optimization method is characterized by being applied to network drop and pull transport of a highway port consisting of a tractor, a trailer and the highway port, and comprising the following steps of:

(1) establishing a network throwing and hanging scheduling target optimization model;

(2) solving the target optimization model in the step (1) by using a two-stage heuristic algorithm:

firstly, the throwing and hanging requirements q between any two highway ports _ij Divided into two fractions q' _ij And q' _ij (ii) a If q is _ij ≥q _ji Then q' _ij ＝q _ji ＝q' _ji ，q” _ij ＝q _ij -q _ji ，q” _ji 0; if q is _ij ＜q _ji Then q' _ij ＝q _ij ＝q' _ji ，q” _ij ＝0，q″ _ji ＝q _ji -q _ij ，q′ _ij The formed matrix is a symmetric matrix; wherein q is _ij : representing the drop and pull transportation requirements of the highway ports i to j by using the drop and pull transportation times; q. q.s _ji : the drop and pull transportation requirements of the highway ports j to i are represented by drop and pull transportation times; q since the requirements are not necessarily balanced _ij Is not necessarily equal to q _ji ；

Stage 1: to q 'of demand' _ij And dispatching the throwing hanging requirement: scheduling the trailer in a throwing manner by adopting two ends of a line, namely selecting a highway port arbitrarily in the network, starting to assign a throwing and hanging task of the tractor, and throwing and hanging the tractor back and forth between the two highway ports with heavy load;

and (2) stage: for q ″' after stage 1 scheduling _ij And dispatching the drop hanging requirement: adopting full-network circulation throwing and hanging vehicle dispatching, namely selecting the least one route in the remaining throwing and hanging requirements after the dispatching of the 1 st stage, firstly throwing and hanging the route at the stage, and taking the starting point of singular requirements as the starting point of the tractor;

the network throwing and hanging scheduling target optimization model in the step (1) is as follows:

an objective function:

the network operation cost is composed of heavy-hanging running cost, empty running cost and fixed cost, and the heavy-hanging running cost

The empty running cost is as follows:

the fixed cost of using the vehicle on the same day is Z _{Fixing device} ＝Kc；

The objective is to express the network operation cost minimum as:

the constraint conditions are as follows:

get rid of and hang the demand and obtain satisfying:

tractor continuous operating time limit:

in the formula, i, j, l: the number of the road port, i, j, l belongs to N;

k: representing the set of tractors, K ═ {1,2, …, K }, and K also represents the total number of tractors required for the road network;

k: representing the number of the tractor, and K belongs to K;

d _ij : represents the distance between the road ports i, j, in units: km;

q _ij : representing the drop and pull transportation requirements of the highway ports i to j by using the drop and pull transportation times;

C _ij ,c _ij : respectively representing the cost of towing, heavy hanging and empty running of tractors from i to j of highway harbors, the unit is as follows: yuan/km;

c: the fixed cost of tractor use in the unit time is shown, including vehicle depreciation, staff wage, the unit: yuan;

t: the time of the continuous work of the tractors is represented, and the time of the continuous work of all the tractors every day is regulated to be equal;

v ₁ ，v ₂ : respectively representing the speed of the tractor for towing heavy hanging and empty running, unit: km/h;

k ₁ ，k ₂ : respectively representing the states of heavy hanging and empty driving of the tractor, wherein K belongs to K;

the number of trips the tractor takes to travel from a highway port to a highway port for a heavy load and an empty ride, respectively.

2. The optimization method for network drop-off transport scheduling according to claim 1, wherein the stage 1 of the step (2) is performed according to the following steps:

the first step is as follows: setting initial parameters, i is 1, K is 0, and Z _{Heavy load} ＝0，Z _{Air conditioner} ＝0，Z _{Fixing device} ＝0；

The second step is that: selecting q' _ij > 0, and max (d) _ij ) The corresponding road port i, j is a throwing and hanging terminal point served by the first vehicle; if q' _ij > 0 and d _ij ≥v ₁ T, then K ═ K +2q' _ij ,q' _ij ＝0＝q' _ji ，Z _{Heavy load} ＝Z _{Heavy load} +2q' _ij d _ij C _ij ，Z _{Fixing device} Kc; if q' _ij > 0 and d _ij ＜v ₁ T, and if the T time can complete the traction pass is

Then K is K +1, Z _{Fixing device} ＝Kc，q' _ij ＝0＝q' _ji And if q' _ij If the working time is more than 0, the departure distance of the highway port j is selected to be d times less than the distance of the highway port j _jl Is taken as the next hanging point until the working time is used up, wherein d _jl Represents the distance from road port j to road port l; corresponding q " _ij ＝q” _ij -1，Z _{Heavy load} ＝Z _{Heavy load} +2q' _ij d _ij C _ij +d _ij C _ij +d _jl C _jl +..; if q " _ij If 0, selecting the departure distance of the highway harbor i to be smaller than d _il Is taken as the next drop and hang point until the vehicle working time is over, wherein d _il Represents the distance from road port i to road port l, corresponding to q' _jl ＝q' _jl -1，Z _{Heavy load} ＝Z _{Heavy load} +2q' _ij d _ij C _ij +d _il C _il +..; if q' _ij > 0 and d _ij ＜v ₁ T, and if

Other vehicles need to be arranged between the highway ports i, j for transportation, and the number of the vehicles is required to be equal to

Q 'if vehicles with residual capacity can still finish other throwing and hanging tasks' _ij ＝0＝q' _ji ，

Z _{Heavy load} ＝Z _{Heavy load} +2q' _ij d _ij C _ij ，Z _{Fixing device} ＝Kc；

The third step: for all i ∈ N, i ═ i +1, if q' _ij If the value is more than 0, the calculation is returned to the second step; if all q' _ij When it is 0, the calculation is stopped, Z ₁ '＝Z _{Heavy load} +Z _{Fixing device} 。

3. The optimization method for network drop-off transport scheduling according to claim 1, wherein the stage 2 of the step (2) is performed as follows:

the first step is as follows: optionally i ∈ N;

the second step is that: q is selected " _ij The highway port i, j greater than 0 is a throwing and hanging scheduling demand point; if d is _ij ≥v ₁ T, then K ═ K + q " _ij ,q” _ij ＝0，Z _{Heavy load} ＝Z _{Heavy load} +q” _ij d _ij C _ij ，Z _{Fixing device} ＝Kc；

If d is _ij ＜v ₁ T, there are two strategies at this time:

strategy 1 is that a vehicle k travels back and forth between highway ports i and j, and only one way of loading goods is needed, and the highway ports j to i are empty; if vehicle k can complete q within time T " _ij The hanging amount is K +1, q " _ij ＝0，Z _{Heavy load} ＝Z _{Heavy load} +q” _ij d _ij C _ij ，Z _{Air conditioner} ＝Z _{Air conditioner} +(q” _ij -1)d _ji c _ji ，Z _{Fixing device} Selecting a departure q in a highway harbor j if the working time is left " _jl Taking the port more than 0 as the next hanging point; if vehicle k cannot complete q within time T " _ij And if the load is thrown and hung, other vehicles are required to be dispatched to throw and hang, and the number of the vehicles required for completing the throwing and hanging task between the highway ports i and j is equal to that of the vehicles

If the vehicle with the residual capacity can still finish other throwing and hanging tasks

q″ _ij ＝0，Z _{Heavy load} ＝Z _{Heavy load} +q” _ij d _ij C _ij ，Z _{Air conditioner} ＝Z _{Air conditioner} +(q” _ij -1)d _ji c _ji ，Z _{Fixing device} ＝Kc；

The strategy 2 is that the vehicle k completes the throwing and hanging task in the highway port network, at the moment, heavy load and no load exist in the whole network, the vehicle k completes one-time throwing and hanging transportation from the highway ports i to j, and q' _ij ＝q” _ij -1, selecting a highway harbor j departure q " _jl L harbor > 0 is used as the next hanging point, q " _jl ＝q” _jl -1, until no working time remains, K ═ K +1, Z _{Heavy load} ＝Z _{Heavy load} +d _ij C _ij +d _jl C _jl +. cndot; if starting from the port j of the highwayAll q " _jl And (5) idling to a port I, selecting a road port with a throwing and hanging task from the port I for throwing and hanging until the working time is not left, wherein K is K +1, and Z _{Heavy load} ＝Z _{Heavy load} +d _ij C _ij +···，Z _{Air conditioner} ＝Z _{Air conditioner} +d _jl c _jl +···，Z _{Fixing device} ＝Kc；

The third step: for all i e N, i +1, if q " _ij If the value is more than 0, the calculation is returned to the second step; if all q " _ij When it is 0, the calculation is stopped, Z ₁ ”＝Z _{Heavy load} +Z _{Air conditioner} +Z _{Fixing device} ；

Therefore, the total cost of the network drop scheduling is as follows: z ₁ ＝Z' ₁ +Z ₁ ”。

4. A network drop and pull transport scheduling optimization method is characterized by being applied to network drop and pull transport of a highway port consisting of a tractor, a trailer and the highway port, and comprising the following steps of:

(1) establishing a network throwing and hanging scheduling target optimization model;

(2) solving the target optimization model in the step (1) by using a heuristic algorithm: the pure network circular throwing and hanging is adopted to throw and hang the demand q of throwing and hanging between any two highway ports _ij Carrying out vehicle dispatching;

the network throwing and hanging scheduling target optimization model in the step (1) is as follows:

an objective function:

the network operation cost is composed of heavy-hanging running cost, empty running cost and fixed cost, and the heavy-hanging running cost

The empty running cost is as follows:

the fixed cost of using the vehicle on the same day is Z _{Fixing device} ＝Kc；

The objective is to express the network operation cost minimum as:

the constraint conditions are as follows:

get rid of and hang the demand and obtain satisfying:

tractor continuous operating time limit:

in the formula, i, j, l: the number of the road port, i, j, l belongs to N;

k: representing the set of tractors, K ═ {1,2, …, K }, and K also represents the total number of tractors required for the road network;

k: representing the number of the tractor, wherein K belongs to K;

d _ij : represents the distance between the road ports i, j, in units: km;

q _ij : the drop and pull transportation requirements of the highway ports i to j are represented by drop and pull transportation times;

C _ij ,c _ij : respectively representing the cost of towing heavy hanging and empty running of tractors from i to j of highway ports, unit: yuan/km;

c: the fixed cost of tractor use in unit time is shown, including vehicle depreciation, staff wage, unit: element;

t: the time of the continuous work of the tractors is represented, and the time of the continuous work of all the tractors every day is regulated to be equal;

v ₁ ，v ₂ : respectively representing the speed of the tractor for towing heavy hanging and empty running, unit: km/h;

k ₁ ，k ₂ : respectively indicating towing heavy-hanging and empty running of tractorK ∈ K;

the number of trips the tractor takes to travel from a highway port to a highway port for a heavy load and an empty ride, respectively.

5. The method for optimizing network drop-off transport scheduling as claimed in claim 4, wherein the step (2) of using pure network loop drop-off for drop-off demand q between any two highway ports _ij The vehicle dispatching is carried out according to the following steps:

the first step is as follows: optionally i ∈ N;

the second step is that: select q _ij The highway port i, j greater than 0 is a throwing and hanging scheduling demand point; if d is _ij ≥v ₁ T, then K ═ K + q _ij ,q _ij ＝0，Z _{Heavy load} ＝Z _{Heavy load} +q _ij d _ij C _ij ，Z _{Fixing device} Kc; if d is _ij ＜v ₁ T, completing a throwing and hanging task by the vehicle k in the highway port network, wherein at the moment, heavy load and no load exist in the whole network, completing one-time throwing and hanging transportation from the highway ports i to j by the vehicle k, and q _ij ＝q _ij -1, selecting the departure q of the port j of the highway _jl L port more than 0 is used as the next hanging point, q _jl ＝q _jl -1, until no working time remains, K-K +1, Z _{Heavy load} ＝Z _{Heavy load} +d _ij C _ij +d _jl C _jl +. cndot; if all q start from the port j of the highway _jl And (5) idling to one port, selecting a road port with a throwing and hanging task from the port I for throwing and hanging until no working time is left, and KK+1，Z _{Heavy load} ＝Z _{Heavy load} +d _ij C _ij +···，Z _{Air conditioner} ＝Z _{Air conditioner} +d _jl c _jl +···，Z _{Fixing device} ＝Kc；

The third step: for all i e N, i +1, if q _ij If the value is more than 0, the calculation is returned to the second step; if all q are _ij When it is 0, the calculation is stopped, Z ₁ ＝Z _{Heavy load} +Z _{Air conditioner} +Z _{Fixing device} ；

Therefore, the net network drop scheduling total cost is: z ₁ ＝Z _{Heavy load} +Z _{Air conditioner} +Z _{Fixing device} 。

Images