CN108550007A - A kind of slotting optimization method and system of pharmacy corporation automatic stereowarehouse - Google Patents

Abstract

The invention discloses a kind of slotting optimization method and system of pharmacy corporation automatic stereowarehouse,The slotting optimization target that this method passes through establishment automatic stereowarehouse,Drug is calculated according to the History Order data of drug in automatic stereowarehouse to go out to be put in storage frequency,According to the correlation degree between every similar drug,Obtain the association factor between every similar drug,Establish piler motion mathematical model,According to piler motion mathematical model,Drug goes out to be put in storage the association factor between frequency and every similar drug,It establishes multiple target slotting optimization mathematical model and multiple target slotting optimization mathematical model is solved,Obtain slotting optimization result,It solves existing method and only considers velocity of goods circulation and shelf stabilities,Practical problem cannot be described well,Cause the problem that optimum results are undesirable,And it is more preferable based on the cargo optimum results that multiple target slotting optimization model solution obtains,Cargo distribution is more reasonable,Substantially increase warehousing and storage activities efficiency,Reduce warehousing operation cost.

Description

A cargo location optimization method and system for an automated three-dimensional warehouse of a pharmaceutical enterprise

technical field

The invention mainly relates to the technical field of logistics storage, in particular to a method and a system for optimizing the location of an automated three-dimensional warehouse of a pharmaceutical enterprise.

Background technique

With the continuous improvement of the overall level of the supply chain in the pharmaceutical industry and the continuous strengthening of industry standards (GMP/GSP), pharmaceutical logistics has become a key area of logistics automation. As an important part of modern logistics, automated three-dimensional warehouse adopts mechanized operation and information scheduling, which has the advantages of saving labor, improving storage management level, and reducing logistics costs. It is an advanced storage mode widely recognized by the society and is widely used in enterprise storage. . The automated three-dimensional warehouses of pharmaceutical companies are connected to the production workshops through aerial conveying lines to realize the full automation of the company's internal logistics from raw materials to finished products, and to solve the buffering needs of products from packaging to off-line to loading. Effective management and control of warehousing costs is one of the most effective means for enterprises to obtain profits.

With the application of electronic label auxiliary system and radio communication transmission technology, the speed and efficiency of picking have been continuously improved, and most of the storage cost is more consumed in the movement of goods in the warehouse. The length of time for goods to move in the warehouse has become one of the key factors to improve the efficiency of warehouse operations. The reasonable layout of goods can effectively reduce the transport distance of transport equipment such as stacker cranes, and reduce the loss of goods during storage and handling. However, due to the continuous entry and exit of goods, some cargo spaces are empty; due to the occurrence of some seasonal diseases such as influenza or sudden infectious diseases, the frequency of goods entering and leaving the warehouse has changed; due to the addition of some goods or some Goods are withdrawn from the market, resulting in crowded or vacant warehouse storage spaces and no storage space for goods. Location optimization is to dynamically reconfigure the location of goods in the warehouse based on changing factors to ensure that the location layout is in a more reasonable state. Therefore, it is of great significance to optimize the location of the warehouse on a regular basis to improve the efficiency of warehousing and reduce the cost of warehousing operations.

The key to cargo location optimization is to establish an optimization model that conforms to the characteristics of the automated three-dimensional warehouse of pharmaceutical companies, and the core of the optimization model is to reasonably establish a mathematical model for multiple optimization objectives, and design the corresponding optimization algorithm based on this. The establishment of mathematical models for multiple optimization objectives relies on the analysis of the particularity of pharmaceuticals in pharmaceutical companies, the characteristics of automated warehouses, and the in-depth analysis of data related to goods, especially the analysis and utilization of historical data. The unit weight of drugs in pharmaceutical companies is relatively light, and there is no need to consider the weight difference between drugs. In the process of cargo location optimization, the running time of the stacker is the main factor affecting the operation efficiency. Moreover, as the seasons change, in different periods, the demanded varieties, demanded quantities and demanded frequency of the goods in the warehouses of pharmaceutical companies will change greatly. Therefore, in combination with the particularity of drugs in pharmaceutical companies, considering the change in the operating speed of the stacker during the optimization process of the storage space, reasonable analysis and utilization of historical data is crucial to design an accurate control algorithm and realize the optimization of storage space in the automated three-dimensional warehouse. premise. In the usual process of cargo space optimization, most of them are based on the turnover rate of the goods and the stability of the shelves to establish a mathematical model for cargo space optimization, and then use the optimization algorithm to optimize the solution. However, the particularity of pharmaceutical companies' drugs was not considered in the process of cargo space optimization, nor was the change in the running speed of the stacker during the process of cargo space optimization, nor was the degree of correlation between drugs considered. Therefore, for this kind of cargo location optimization method, because the particularity of the optimization object is not fully considered, and the historical data of the automated three-dimensional warehouse is not fully utilized, the optimization effect is not ideal, which affects the operational efficiency of the warehouse and the operating cost of the warehouse.

Contents of the invention

The cargo location optimization method and system of the automatic three-dimensional warehouse of pharmaceutical enterprises provided by the present invention solve the problem that the existing method only considers the turnover rate of cargo and shelf stability, and cannot describe the actual problem well, resulting in unsatisfactory optimization results.

In order to solve the above-mentioned technical problems, the cargo location optimization method of the automatic three-dimensional warehouse of pharmaceutical enterprises proposed by the present invention comprises:

The cargo location optimization system of the automatic three-dimensional warehouse of pharmaceutical enterprises proposed by the present invention comprises:

Establish the location optimization goal of the automated three-dimensional warehouse;

Calculate the frequency of drug entry and exit based on the historical order data of drugs in the automated three-dimensional warehouse;

Classify the drugs in the automated three-dimensional warehouse, and obtain the correlation factor between each type of drugs according to the degree of correlation between each type of drugs;

According to the change of the movement speed of the stacker in the process of cargo location optimization, the mathematical model of the movement of the stacker is established;

Based on the mathematical model of stacker movement, the frequency of drug entry and exit, and the correlation factors between each type of drug, a multi-objective cargo space optimization mathematical model is established;

Solve the multi-objective cargo location optimization mathematical model to obtain the cargo location optimization results.

Further, the goal of optimizing the location of the automated three-dimensional warehouse is established as follows:

Establish the frequency of in-out and out-of-stock medicines in the automated three-dimensional warehouse and the relevance of the medicines as the goal of optimizing the location of the automated three-dimensional warehouse.

Further, the calculation formula for calculating the frequency of medicines entering and leaving the warehouse based on the historical order data of medicines in the automated three-dimensional warehouse is:

Among them, p _ij is the stock-in and out frequency of the jth drug in the i-th category, M _ij is the total number of stock-in and out of the j-th drug in the i-th category in the corresponding production cycle, and S is the total number of all drugs in and out in the corresponding production cycle.

Further, classify the drugs in the automated three-dimensional warehouse, and according to the degree of correlation between each type of drugs, get the correlation factors between each type of drugs including:

Divide all the medicines in the automated three-dimensional warehouse into n categories, of which there are k _i medicines in the i category;

Calculate the correlation factor between each class of Chinese medicines as Among them, r _ifg is the correlation factor of two drugs f,g in the i-th category, f,g(f,g∈1,2,...,j,...k _i ), Q is the corresponding production cycle The total number of orders, q _ifg is the number of orders that include drugs f and g at the same time;

According to the correlation factors between each class of Chinese medicines, the correlation matrix of the correlation factors of similar drugs is obtained as follows:

Obtain the drug e with the strongest correlation with other drugs in category i according to the correlation matrix;

Calculate the correlation factor of other drugs in category i relative to drug e, denoted as

Further, according to the change of the movement speed of the stacker during the optimization process of the cargo space, the establishment of the mathematical model of the movement of the stacker includes:

Calculate the time spent in the horizontal direction when the stacker picks the jth drug of the i category:

s _xij = x _ij ^l

Among them, t _xij is the time spent in the horizontal direction when the stacker picks the jth drug of the i category, and s _xij is the moving distance in the horizontal direction when the stacker picks the jth drug of the i category, a _x is the acceleration of the stacker in the horizontal direction, v _xmax is the maximum running speed of the stacker in the horizontal direction, S _xmax is the maximum horizontal running distance of the stacker when it accelerates uniformly to v _xmax , x _ij is the i-th category The coordinate value of the jth drug in the horizontal direction, l is the length of the cargo box;

Calculate the time spent in the vertical direction when the stacker picks the jth drug of the i category:

s _zij =z _ij ^h

Among them, t _zij is the time spent in the vertical direction when the stacker picks the jth drug of the i category, and s _zij is the moving distance in the vertical direction when the stacker picks the jth drug of the i category, a _z is the acceleration of the stacker in the vertical direction, v _zmax is the maximum running speed of the stacker in the vertical direction, S _zmax is the maximum vertical running distance of the stacker when it accelerates uniformly to v _zmax , z _ij is the i-type The coordinate value of the jth medicine in the vertical direction, h is the height of the cargo box;

The time it takes to get the stacker to pick the jth medicine of the i category is:

t _ij =max(t _xij ,t _zij ).

Further, according to the mathematical model of the movement of the stacker, the frequency of entering and leaving the warehouse of medicines, and the correlation factors between each type of medicines, a multi-objective cargo space optimization mathematical model is established as follows:

at the same time,

And i, j, x _ij , y _ij , z _ij are all integers,

Among them, n is the number of categories of medicines in the automated three-dimensional warehouse, _ki is the number of medicines in the i-th category of medicines, p _ij is the frequency of entry and exit of the j-th medicine in the i-th category, and t _ij is the number of the i-th category picked by the stacker. The time spent on the jth drug, (x _ij , y _ij , z _ij ) is the optimized position of the jth drug in the i category, v _y is the uniform transport speed of the conveyor belt, It is the optimized position of the drug in category i that has the strongest correlation with other drugs in this category, r and w are divided into the distance between other lanes and the width of the cargo box, is the average position of all medicines, and A, B, and C are respectively the maximum number of columns, the maximum number of rows, and the maximum number of floors of the warehouse.

Further, solving the multi-objective cargo location optimization mathematical model is specifically: using the NSGA-II algorithm to solve the multi-objective cargo location optimization mathematical model.

Further, using the NSGA-II algorithm to solve the multi-objective cargo location optimization mathematical model includes:

Integer encoding is used to encode the position of the drug in the automated three-dimensional warehouse. The population is a matrix, and each row is a chromosome, corresponding to a feasible solution. Set the population number, maximum evolutionary algebra, crossover probability and mutation probability;

Using the Mersenne Twister algorithm, the initial population is randomly generated, and the reciprocal of the objective function in the multi-objective cargo space optimization mathematical model is used as the fitness function to calculate the individual fitness value, and perform fast non-dominated sorting (Non-Dominated) Sort), calculate the degree of congestion (Crowding Distance);

Form a new population by using the binary tournament selection strategy (Tournament Selection);

Use the simulated binary crossover operator (SBX) and the polynomial mutation operator (polynomial mutation) to perform crossover and mutation operations respectively to generate offspring populations;

Merge the parent population and the child population into a temporary population, then perform non-dominated sorting, calculate the congestion distance, and use the congestion comparison operator to select a new parent population;

On this basis, perform selection, crossover, and mutation operations to form a new offspring population. If the current evolutionary algebra is greater than the maximum evolutionary algebra, stop evolution and obtain a set of Pareto optimal solution sets.

The cargo location optimization system of the automatic three-dimensional warehouse of pharmaceutical enterprises provided by the present invention comprises:

A memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the computer program, the steps of the above method for optimizing the location of the automated three-dimensional warehouse of a pharmaceutical enterprise are realized.

Compared with the prior art, the present invention has the advantages of:

The cargo location optimization method and system of the automated three-dimensional warehouse of pharmaceutical enterprises provided by the present invention, by establishing the cargo location optimization target of the automated three-dimensional warehouse, calculating the frequency of medicines in and out of the warehouse according to the historical order data of medicines in the automated three-dimensional warehouse, for the automated three-dimensional warehouse. Drugs are classified, and according to the degree of correlation between each type of drug, the correlation factor between each type of drug is obtained. According to the change of the movement speed of the stacker during the optimization process of the cargo space, a mathematical model of the stacker movement is established. According to the stacker The mathematical model of stacker movement, the frequency of drug in and out of the warehouse, and the correlation factors between each type of drug, establish a multi-objective cargo location optimization mathematical model and solve the multi-objective cargo location optimization mathematical model to obtain the cargo location optimization result, which solves the existing problems. The method only considers the turnover rate of goods and shelf stability, which cannot describe the actual problem well, resulting in unsatisfactory optimization results, and the established multi-objective optimization mathematical model fully considers the particularity of drugs in pharmaceutical companies, and the automated three-dimensional warehouse The characteristics and actual working conditions, so that the cargo optimization results obtained by solving the multi-objective cargo location optimization mathematical model are more ideal, and the cargo distribution is more reasonable, which greatly improves the efficiency of storage operations and reduces the cost of storage operations.

Description of drawings

Fig. 1 is the flow chart of the goods position optimization method of the automatic three-dimensional warehouse of the pharmaceutical enterprise of embodiment one of the present invention;

Fig. 2 is the flow chart of the cargo location optimization method of the automatic three-dimensional warehouse of the pharmaceutical enterprise in the second embodiment of the present invention;

Fig. 3 is an algorithm flow chart for solving the multi-objective cargo location optimization mathematical model by using the NSGA-II algorithm in Embodiment 2 of the present invention;

Fig. 4 is the cargo location distribution diagram before drug optimization in the cargo location optimization method of the automated three-dimensional warehouse of the pharmaceutical enterprise in the second embodiment of the present invention;

Fig. 5 is the cargo location distribution diagram after the optimization of medicines in the cargo location optimization method of the automated three-dimensional warehouse of the pharmaceutical enterprise in the second embodiment of the present invention;

Fig. 6 is a structural block diagram of a cargo location optimization system for an automated three-dimensional warehouse of a pharmaceutical enterprise according to an embodiment of the present invention.

Reference signs:

10. Memory; 20. Processor.

Detailed ways

In order to facilitate the understanding of the present invention, the present invention will be described more fully and in detail below in conjunction with the accompanying drawings and preferred embodiments, but the protection scope of the present invention is not limited to the following specific embodiments.

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways defined and covered by the claims.

Embodiment one

Referring to Fig. 1 , the cargo location optimization method of an automated three-dimensional warehouse of a pharmaceutical enterprise provided by Embodiment 1 of the present invention includes:

Step S101, establishing the cargo location optimization goal of the automated three-dimensional warehouse;

Step S102, calculating the frequency of medicines entering and leaving the warehouse according to the historical order data of medicines in the automated three-dimensional warehouse;

Step S103, classify the drugs in the automated three-dimensional warehouse, and obtain the correlation factor between each type of drugs according to the degree of correlation between each type of drugs;

Step S104, according to the change of the movement speed of the stacker during the cargo location optimization process, a mathematical model of the movement of the stacker is established;

Step S105, according to the mathematical model of the movement of the stacker, the frequency of medicines entering and leaving the warehouse, and the correlation factors between each type of medicines, a multi-objective cargo space optimization mathematical model is established;

Step S106, solving the multi-objective mathematical model for cargo location optimization to obtain a cargo location optimization result.

The cargo location optimization method of the automated three-dimensional warehouse of the pharmaceutical enterprise provided by the embodiment of the present invention, by establishing the cargo location optimization target of the automated three-dimensional warehouse, calculating the frequency of medicines entering and leaving the warehouse according to the historical order data of the medicines in the automated three-dimensional warehouse, for the automated three-dimensional warehouse. Drugs are classified, and according to the degree of correlation between each type of drug, the correlation factor between each type of drug is obtained. According to the change of the movement speed of the stacker during the optimization process of the cargo space, a mathematical model of the stacker movement is established. According to the stacker The mathematical model of stacker movement, the frequency of drug in and out of the warehouse, and the correlation factors between each type of drug, establish a multi-objective cargo location optimization mathematical model and solve the multi-objective cargo location optimization mathematical model to obtain the cargo location optimization result, which solves the existing problems. The method only considers the turnover rate of goods and shelf stability, which cannot describe the actual problem well, resulting in unsatisfactory optimization results, and the established multi-objective optimization mathematical model fully considers the particularity of drugs in pharmaceutical companies, and the automated three-dimensional warehouse The characteristics and actual working conditions, so that the cargo optimization results obtained by solving the multi-objective cargo location optimization mathematical model are more ideal, and the cargo distribution is more reasonable, which greatly improves the efficiency of storage operations and reduces the cost of storage operations.

Specifically, according to the requirements of pharmaceutical companies, the embodiment of the present invention establishes the cargo location optimization goal of the automated three-dimensional warehouse in combination with the particularity of the drugs of the pharmaceutical companies. The specificity of the drugs of pharmaceutical companies is manifested in the following aspects:

①According to the requirements of GMP/GSP standards, drugs need to be classified and stored according to different natural attributes.

②The unit weight of medicines is mostly small, and the weight of medicines may not be considered when optimizing the storage space.

③ There is a strong correlation between medicines. In order to ensure the access efficiency, the medicines with strong correlation should be placed together.

Therefore, when establishing the optimization goal, first of all, the medicines with high frequency of entry and exit should be placed closer to the entrance and exit, so as to shorten the time it takes for the stacker to pick goods and improve the efficiency of storage operations. At the same time, weight differences between medicines are no longer considered, that is, shelf stability is no longer considered. Then, on the basis of classified storage of drugs, analyze the correlation degree of all drugs in each category, so that the drugs with strong correlation degree within the category are placed together, so as to shorten the time spent on picking the same batch of inbound and outbound goods, and further improve storage efficiency. High operating efficiency and reduced storage operating costs. As a result, the frequency of medicines entering and leaving the warehouse and the relevance of medicines are established as the optimization goals.

Embodiment two

Referring to Fig. 2 , the cargo location optimization method for the automated three-dimensional warehouse of a pharmaceutical enterprise provided by Embodiment 2 of the present invention includes:

Step S201, establishing the relationship between the frequency of medicines entering and leaving the warehouse in the automated warehouse and the correlation between medicines as the cargo location optimization goal of the automated warehouse.

Step S202, calculating the frequency of medicines entering and leaving the warehouse according to the historical order data of medicines in the automated three-dimensional warehouse.

Specifically, within a production cycle, the frequency of entry and exit of the jth drug in category i is:

Among them, p _ij is the stock-in and out frequency of the jth drug in the i-th category, M _ij is the total number of stock-in and out of the j-th drug in the i-th category in the corresponding production cycle, and S is the total number of all drugs in and out in the corresponding production cycle.

Step S203, classify the drugs in the automated three-dimensional warehouse, and obtain the correlation factor between each type of drugs according to the degree of correlation between each type of drugs.

Specifically, according to the principle of classification and storage of medicines, the medicines in the automated three-dimensional warehouse are classified. Divide all medicines into n categories, and there are k _i medicines in category i. For the j-th drug of category i, the optimized position is (x _ij , y _ij , z _ij ). In the same production cycle, the number of orders in which two drugs f,g (f,g∈1,2,...,j,...k _i ) appear in the same order in the i-th category accounts for the total number of orders The ratio of is the support of association rules. For the convenience of calculation and expression, the correlation factor of two drugs f and g defined as the i-th category of support is:

Wherein, at the same time, the total number of orders in the corresponding production cycle is Q, and the number of orders including drugs f and g is q _ifg .

Thus, the correlation factors among all drugs in the i-th category within the time period, the correlation matrix of the correlation factors of similar drugs is obtained:

The drug e with the strongest correlation with other drugs in category i is selected from the correlation matrix, and its optimized position is denoted as

Thus, the correlation factor of other drugs in category i relative to drug e is further obtained, denoted as

Step S204, according to the change of the movement speed of the stacker during the cargo location optimization process, a mathematical model of the movement of the stacker is established.

Specifically, this embodiment analyzes the change of the movement speed of the stacker in the process of cargo location optimization according to the actual working conditions, abstracts the movement process of picking a cargo by the stacker into a process of uniform acceleration and deceleration, and establishes the The mathematical model of motion is:

When the stacker picks the j-th drug of the i-th category, the time spent in the horizontal direction is:

s _xij = x _ij ^l (5)

Among them, t _xij is the time spent in the horizontal direction when the stacker picks the jth drug of the i category, and s _xij is the moving distance in the horizontal direction when the stacker picks the jth drug of the i category, a _x is the acceleration of the stacker in the horizontal direction, v _xmax is the maximum running speed of the stacker in the horizontal direction, S _xmax is the maximum horizontal running distance of the stacker when it accelerates uniformly to v _xmax , x _ij is the i-th category The coordinate value of the jth drug in the horizontal direction, l is the length of the cargo box;

Similarly, when the stacker picks the jth medicine of the i-th category, the time spent in the vertical direction is:

s _zij =z _ij ^h (8)

Among them, t _zij is the time spent in the vertical direction when the stacker picks the jth drug of the i category, and s _zij is the moving distance in the vertical direction when the stacker picks the jth drug of the i category, a _z is the acceleration of the stacker in the vertical direction, v _zmax is the maximum running speed of the stacker in the vertical direction, S _zmax is the maximum vertical running distance of the stacker when it accelerates uniformly to v _zmax , z _ij is the i-type The coordinate value of the jth medicine in the vertical direction, h is the height of the cargo box;

Then, the time it takes for the stacker to pick the j-th drug of the i-th category is:

t _ij ＝max(t _xij ,t _zij ) (9)

In step S205, a multi-objective mathematical model for cargo space optimization is established according to the mathematical model of the movement of the stacker, the frequency of medicines entering and leaving the warehouse, and the correlation factors between each type of medicines.

In order to realize the nearby storage of medicines, the medicines with high frequency of entry and exit are stored in a place close to the warehouse entry and exit platform, so that the sum of the frequency of entry and exit of medicines and the time spent on picking the medicines is the smallest, and it is assumed that each shipment Only one drug is stored in the grid, and the time for the goods to go from the stacker to the conveyor belt is ignored, that is

Among them, n is the number of categories of medicines in the automated three-dimensional warehouse, k _i is the number of medicines of the i-th category of medicines, p _ij is the frequency of entry and exit of the j-th medicine of the i-th category, and t _ij is the number of medicines picked by the stacker The time spent on the jth drug of class i, (x _ij , y _ij , z _ij ) is the optimized position of the jth drug of class i, v _y is the uniform transport speed of the conveyor belt, r, w are respectively The distance between the lanes and the width of the cargo grid.

According to the strong correlation between medicines, on the basis of classification and storage of medicines, according to historical data, association analysis method is used to analyze the degree of correlation between all medicines in each category, so that the medicines with strong intra-category correlation Put them together to shorten the time it takes to pick the same batch of inbound and outbound goods.

For all drugs in category i, take drug e as the center position, calculate the Euclidean distance between other drugs of this type and drug e, and use the correlation factor obtained above as the correlation weight to make the Euclidean distance between other drugs of this type and drug e The sum of the products with the corresponding associated weights is the smallest, that is:

In order to make all the medicines in the warehouse as close as possible to the entrance and exit, the Euclidean distance between the average position of all medicines and the entrance and exit is the smallest, namely:

and then,

in, is the optimized position of the drug in category i that has the strongest correlation with other drugs in this category, Represents the average position of all medicines, and d represents the Euclidean distance from the average position of all medicines to the warehouse entrance.

At the same time, in order to store each type of medicine as scattered as possible, the sum of the Euclidean distances between the center position of each type of medicine and the average position of all medicines is maximized, namely:

From this, we get:

In summary, the mathematical model of cargo space optimization with multiple optimization objectives is obtained:

at the same time,

And i, j, x _ij , y _ij , z _ij are all integers,

Among them, A, B, and C are respectively the maximum number of columns, the maximum number of rows, and the maximum number of layers of the warehouse.

In step S206, the NSGA-II algorithm is used to solve the multi-objective cargo space optimization mathematical model to obtain a cargo space optimization result.

With reference to Fig. 3, Fig. 3 adopts NSGA-II algorithm to solve the algorithm flow chart of multi-objective cargo space optimization mathematical model for the embodiment of the present invention, and its specific steps are:

①Integer encoding is used to encode the location of the drug. The population is a matrix, and each row is a chromosome, corresponding to a feasible solution. Set the population size, maximum evolutionary generation, crossover probability and mutation probability.

The information of each drug includes the code of the drug, the code of the storage location where it is located, the number of the storage location, and the like. Due to the mapping relationship between the drug and its location, the cargo code of each drug is fixed, while its storage location code and location number change dynamically with the movement of the drug. Therefore, the present invention adopts integer coding, and selects the cargo location where the drug is located as the gene on the chromosome:

1) The population is a matrix, each row is a chromosome, corresponding to a feasible solution, that is, corresponding to a cargo location optimization scheme;

2) The number of genes contained in each chromosome represents the number of goods to be optimized;

3) Each gene on each chromosome represents the location information of a drug, respectively representing the column, row, and row of the drug. That is, each gene is represented by 3 integers.

As shown in Table 1:

Table 1 Chromosome coding

As shown in Table 1, 153 stands for (1, 5, 3), indicating that the drug with the cargo location number 1 is stored in the first column, fifth row, third row.

②Adopt the Mersenne Twister algorithm (Mersenne Twister) to randomly generate the initial population. The reciprocal of the objective function is used as the fitness function to calculate the individual fitness value, and perform a fast non-dominated sort (Non-Dominated Sort) to calculate the crowding distance (Crowding Distance).

In genetic algorithm, fitness is used to measure the degree to which individuals in the population reach or approach the optimal solution in the optimization calculation. Individuals with higher fitness have a higher probability of passing on to the next generation; while individuals with lower fitness have a relatively smaller probability of passing on to the next generation. The main basis of the genetic algorithm to guide the search is the fitness value of the individual. That is, genetic algorithms rely on selection operations to guide the algorithm's search direction. The selection operation uses the fitness value of the individual as a deterministic index, and selects individuals with high fitness values from the current population for crossover and mutation to find the optimal solution. The three objective functions of this embodiment all seek the global minimum value, therefore, take the reciprocal of the objective function as the fitness function, then the corresponding fitness function after the objective function conversion is:

Among them, F ₁ , F ₂ , and F ₃ respectively correspond to the objective functions in formula (16).

③Using the binary tournament selection strategy (Tournament Selection) to form a new population.

④ Use simulated binary crossover operator (SBX) and polynomial mutation operator (polynomial mutation) to perform crossover and mutation operations respectively to generate offspring populations.

⑤ Merge the parent population and the offspring population into a temporary population, then perform non-dominated sorting, and calculate the crowding distance. Use the crowding degree comparison operator to select a new parent population.

⑥On this basis, carry out selection, crossover and mutation operations to form new offspring populations. If the current evolution algebra is greater than the maximum evolution algebra, then stop the evolution and get a set of Pareto optimal solutions.

An example is given below to describe the location optimization method of the present invention in detail:

Table 2 shows the working condition parameters of the finished product warehouse of the automated three-dimensional warehouse of the pharmaceutical company.

Table 2 Example simulation parameter information table

Simulation parameters value Simulation parameters value The maximum operating speed of the stacker in the horizontal direction m/s 2.0 The maximum number of columns of three-dimensional shelves 30 The maximum running speed of the stacker in the vertical direction m/s 0.6 The maximum number of rows of three-dimensional shelves 20 Stacker horizontal acceleration m/S ² 0.3 The maximum number of layers of three-dimensional shelves 10 Stacker vertical acceleration m/S ² 0.5 The position coordinates of the warehouse entry and exit (0,0,0) Conveyor belt running speed m/s 0.5 Population number 100 The length of each compartment m 1.0 maximum evolution algebra 500 Width of each compartment m 1.2 crossover probability 0.8 The height of each compartment m 1.5 mutation probability 0.1

The known initial data of the drugs to be optimized are shown in Table 3.

Table 3 Known initial data of drugs to be optimized

Using the cargo location optimization model proposed by formula (1) ~ formula (16), optimize the cargo location of the goods in the warehouse, and use MATLAB software for simulation.

Adopt the method for the second embodiment of the present invention to carry out the cargo location optimization of the automatic three-dimensional warehouse of the pharmaceutical enterprise, the cargo location positions before and after the optimization of the medicine are as shown in Figure 4 and Figure 5, wherein represents class 1 drugs, represents class 2 drugs, Represents class 3 drugs. From Figure 4 and Figure 5, it can be clearly seen that the optimized drug storage location is more reasonable, the same type of drugs are stored together, and different types of drugs are stored separately, and the optimized drug storage location is also closer to the warehouse as a whole.

It can be seen that the cargo location optimization method for the automated three-dimensional warehouse of pharmaceutical companies provided by the embodiment of the present invention makes the distribution of goods more reasonable, and solves the problem that the existing method only considers the turnover rate of goods and shelf stability, and cannot describe the actual problem well. The problem of unsatisfactory optimization results, and the established multi-objective location optimization mathematical model fully considers the particularity of pharmaceutical companies, the characteristics of automated warehouses and actual working conditions, so that the solution based on the multi-objective location optimization mathematical model can be obtained The goods optimization results are more ideal, and the goods distribution is more reasonable, which greatly improves the efficiency of warehousing operations and reduces the cost of warehousing operations.

Referring to Fig. 6, the cargo location optimization system of the automated three-dimensional warehouse of pharmaceutical enterprises proposed by the embodiment of the present invention includes:

The memory 10, the processor 20, and the computer program stored on the memory 10 and operable on the processor 20, wherein, when the processor 20 executes the computer program, realizes the cargo location optimization method of the pharmaceutical enterprise automated three-dimensional warehouse of the embodiment of the present invention step.

For the specific working process and working principle of the storage location optimization system of the automated three-dimensional warehouse of pharmaceutical enterprises in this embodiment, please refer to the working process and working principle of the storage location optimization method of the automatic three-dimensional warehouse of pharmaceutical enterprises in this embodiment.

The above are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (9)

1. a cargo location optimization method of a pharmaceutical enterprise automated three-dimensional warehouse, characterized in that, the method comprises: Establish the location optimization goal of the automated three-dimensional warehouse; According to the historical order data of medicines in the automated three-dimensional warehouse, the frequency of medicines entering and leaving the warehouse is calculated; Classify the drugs in the automated three-dimensional warehouse, and obtain the correlation factor between each type of drugs according to the degree of correlation between each type of drugs; According to the change of the movement speed of the stacker in the process of cargo location optimization, the mathematical model of the movement of the stacker is established; Establish a multi-objective cargo space optimization mathematical model according to the mathematical model of the movement of the stacker, the frequency of entering and leaving the warehouse of the medicines, and the correlation factors between each type of medicines; The multi-objective cargo space optimization mathematical model is solved to obtain a cargo space optimization result. 2. the cargo location optimization method of the pharmaceutical enterprise automated three-dimensional warehouse according to claim 1, is characterized in that, the cargo location optimization target of establishing the automated three-dimensional warehouse is specifically: Establishing the frequency of entering and exiting medicines in the automated three-dimensional warehouse and the relevance of medicines are the cargo location optimization goals of the automated three-dimensional warehouse. 3. the cargo location optimization method of the pharmaceutical enterprise automated three-dimensional warehouse according to claim 2, is characterized in that, according to the historical order data of medicine in the described automated three-dimensional warehouse, the calculation formula for calculating the frequency of entering and leaving the warehouse of medicines is: Among them, p _ij is the stock-in and out frequency of the jth drug in the i-th category, M _ij is the total number of stock-in and out of the j-th drug in the i-th category in the corresponding production cycle, and S is the total number of all drugs in and out in the corresponding production cycle. 4. The cargo location optimization method of the automated three-dimensional warehouse of a pharmaceutical enterprise according to claim 3, wherein the medicines in the automated three-dimensional warehouse are classified, and according to the degree of association between each type of medicine, each type of medicine is obtained. Correlation factors between drugs include: Divide all medicines in the automated three-dimensional warehouse into n categories, wherein there are k _i medicines in total in the i category; Calculate the correlation factor between each class of Chinese medicines as Among them, r _ifg is the correlation factor of two drugs f,g in the i-th category, f,g(f,g∈1,2,...,j,...k _i ), Q is the corresponding production cycle The total number of orders, q _ifg is the number of orders that include drugs f and g at the same time; According to the correlation factors between each class of Chinese medicines, the correlation matrix of the correlation factors of similar drugs is obtained as follows: Obtain the drug e with the strongest correlation with other drugs in category i according to the correlation matrix; Calculate the correlation factor of other drugs in category i relative to drug e, denoted as 5. according to the cargo position optimization method of the automatic three-dimensional warehouse of the pharmaceutical enterprise described in any one of claim 1-4, it is characterized in that, according to the variation of the motion speed of the stacker in the cargo space optimization process, the mathematical model of the motion of the stacker is established include: Calculate the time spent in the horizontal direction when the stacker picks the jth drug of the i category: s _xij = x _ij ^l Among them, t _xij is the time spent in the horizontal direction when the stacker picks the jth drug of the i category, and s _xij is the moving distance in the horizontal direction when the stacker picks the jth drug of the i category, a _x is the acceleration of the stacker in the horizontal direction, v _xmax is the maximum running speed of the stacker in the horizontal direction, S _xmax is the maximum horizontal running distance of the stacker when it accelerates uniformly to v _xmax , x _ij is the i-th category The coordinate value of the jth drug in the horizontal direction, l is the length of the cargo box; Calculate the time spent in the vertical direction when the stacker picks the jth drug of the i category: s _zij =z _ij h Among them, t _zij is the time spent in the vertical direction when the stacker picks the jth drug of the i category, and s _zij is the moving distance in the vertical direction when the stacker picks the jth drug of the i category, a _z is the acceleration of the stacker in the vertical direction, v _zmax is the maximum running speed of the stacker in the vertical direction, S _zmax is the maximum vertical running distance of the stacker when it accelerates uniformly to v _zmax , z _ij is the i-type The coordinate value of the jth medicine in the vertical direction, h is the height of the cargo box; The time it takes to get the stacker to pick the jth medicine of the i category is: t _ij =max(t _xij ,t _zij ). 6. according to the cargo place optimization method of pharmaceutical enterprise automatic three-dimensional warehouse according to claim 5, it is characterized in that, according to the correlation factor between the kinematics model of stacker, the storage frequency of medicines and every type of medicines, establish multiple The mathematical model of target cargo space optimization is: at the same time, And i, j, x _ij , y _ij , z _ij are all integers, Among them, n is the number of categories of medicines in the automated three-dimensional warehouse, k _i is the number of medicines of the i-th category of medicines, p _ij is the frequency of entry and exit of the j-th medicine of the i-th category, and t _ij is the number of medicines picked by the stacker The time spent on the jth drug in category i, (x _ij , y _ij , z _ij ) is the optimized position of the jth drug in category i, v _y is the uniform transport speed of the conveyor belt, It is the optimized position of the drug in category i that has the strongest correlation with other drugs in this category, r and w are divided into the distance between other lanes and the width of the cargo box, is the average position of all medicines, and A, B, and C are respectively the maximum number of columns, the maximum number of rows, and the maximum number of floors of the warehouse. 7. according to the cargo location optimization method of pharmaceutical enterprise automated three-dimensional warehouse according to claim 6, it is characterized in that, solving described multi-objective cargo location optimization mathematical model is specifically: The NSGA-II algorithm is used to solve the multi-objective cargo location optimization mathematical model. 8. The cargo location optimization method of the automated three-dimensional warehouse of a pharmaceutical enterprise according to claim 7, characterized in that, using the NSGA-II algorithm, solving the multi-objective cargo location optimization mathematical model includes: Integer encoding is used to encode the position of the drug in the automated three-dimensional warehouse. The population is a matrix, and each row is a chromosome, corresponding to a feasible solution, and the number of populations, maximum evolution algebra, crossover probability and mutation probability are set; Using the Mersenne Twister algorithm (Mersenne Twister), the initial population is randomly generated, and the reciprocal of the objective function in the multi-objective cargo space optimization mathematical model is used as the fitness function to calculate the individual fitness value, and perform fast non-dominated sorting (Non -Dominated Sort), calculate the degree of congestion (Crowding Distance); Form a new population by using the binary tournament selection strategy (Tournament Selection); Use the simulated binary crossover operator (SBX) and the polynomial mutation operator (polynomial mutation) to perform crossover and mutation operations respectively to generate offspring populations; Merge the parent population and the child population into a temporary population, then perform non-dominated sorting, calculate the congestion distance, and use the congestion comparison operator to select a new parent population; On this basis, perform selection, crossover, and mutation operations to form a new offspring population. If the current evolutionary algebra is greater than the maximum evolutionary algebra, stop evolution and obtain a set of Pareto optimal solution sets. 9. A cargo location optimization system for an automated three-dimensional warehouse of a pharmaceutical company, characterized in that the system includes: A memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method in any one of claims 1 to 8 when executing the computer program.