Planning device, planning method, and program

Abstract

To provide a planning device, a planning method, and a program which prevent processing planning in which processing timings of objects of the same type are concentrated.SOLUTION: A processing apparatus 110 comprises an acquisition unit 111 which acquires at least prerequisite information about attributes of a plurality of objects as prerequisite information for making a processing plan, an evaluation index calculation unit 113 which calculates a value of an evaluation index for evaluating the processing plan by using the prerequisite information, and a processing planning unit 114 which makes the processing plan on the basis of an evaluation result of the processing plan obtained by using the value of the evaluation index. The evaluation index includes an attribute evaluation index which evaluates degrees of similarity of the attributes of the plurality of objects, and the degrees of similarity of the attributes of the plurality of objects, which are evaluated by the attribute evaluation index include degrees of similarity of attributes of at least partial combinations of pairs of objects which can be selected from objects included in the processing plan.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a plan creation device, a plan creation method, and a program. [Background technology]

[0002] In some cases, it is preferable to create a processing plan for processing objects, such as a production plan or a logistics plan, in which the processing is as uniform as possible. Patent Document 1 describes a technique for creating a processing plan in such a way that the processing is uniform.

[0003] Patent document 1 describes determining the timing for dispatching unallocated ships to loading ports so as to level out the timing of ships' arrival at loading ports and bring the total volume of each brand to be taken during the planning period close to a set volume. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-13238 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology described in Patent Document 1 merely makes the loading capacity of each unallocated ship that is dispatched at equal intervals in time the same, and selects a combination of loading capacity patterns for each brand so that the total volume of each brand approaches a set volume. In other words, the technology described in Patent Document 1 only needs to achieve equal intervals for the timing of dispatching unallocated ships to loading ports and to bring the total volume of each brand of the brand close to a set volume.

[0006] Therefore, in the technology described in Patent Document 1, for example, since ships loaded with the same brand of raw materials are allowed to be dispatched at close times, a processing plan may be created in which the same brand of raw materials arrive at a steelworks at close times. Or, since ships with the same brand of raw materials and similar loading amounts are allowed to be dispatched at close times, a processing plan may be created in which the unloading amount of the same brand of raw materials arriving at the steelworks is biased over time. In this case, assuming that the ships arrive at the unloading port in roughly the order of departure from the loading port, unloading work for the same brand of raw materials will be concentrated at the steelworks (unloading port). As a result, there is a risk of inventory shortages of other brands of raw materials, or overstocking or understocking due to the time bias of inventory amounts within the same brand, and the processing plan may not be sufficient to handle cases where multiple types of raw materials are mixed and used in a specified ratio in blast furnace operation. In addition, even in cases other than blast furnace operation, for example, when a single factory produces a wide variety of products in small quantities and ships them evenly over time, it may be necessary to produce a wide variety of products in a time-distributed manner as much as possible. However, the technology described in Patent Document 1 may create a processing plan in which the processing timing of objects of the same type is concentrated, making it difficult to meet such needs. Therefore, it is desirable to create a processing plan that suppresses the concentration of the processing timing of objects of the same type and processes various objects in a time-distributed manner as much as possible.

[0007] The present invention has been made in consideration of the above problems, and has an object to prevent the creation of a processing plan in which processing timings for objects of the same type are concentrated. [Means for solving the problem]

[0008] The plan creation device of the present invention is a plan creation device that creates a processing plan for processing a plurality of objects, and includes an acquisition unit that acquires, as premise information for creating the processing plan, at least premise information related to attributes of the plurality of objects, an evaluation index calculation unit that uses the premise information to calculate an evaluation index value for evaluating the processing plan, and a processing plan creation unit that creates the processing plan based on an evaluation result of the processing plan using the evaluation index value. the evaluation index includes an attribute evaluation index for evaluating the degree of similarity of attributes of the plurality of objects, and the degree of similarity of attributes of the plurality of objects evaluated by the attribute evaluation index includes the degree of similarity of attributes for at least some combinations of two objects selectable from the objects to be included in the processing plan.

[0009] The planning method of the present invention is a planning method for creating a processing plan for processing a plurality of objects, and comprises an acquisition step in which an acquisition means acquires at least premise information regarding the attributes of the plurality of objects as premise information for creating the processing plan, an evaluation index calculation step in which an evaluation index calculation means uses the premise information to calculate a value of an evaluation index for evaluating the processing plan, and a processing plan creation step in which a processing plan creation means creates the processing plan based on a result of evaluation of the processing plan using the value of the evaluation index, wherein the evaluation index includes an attribute evaluation index for evaluating the degree of similarity of the attributes of the plurality of objects, and the degree of similarity of the attributes of the plurality of objects evaluated by the attribute evaluation index includes the degree of similarity of attributes for at least some combinations of two objects selectable from the objects to be included in the processing plan.

[0010] The program of the present invention causes a computer to function as each unit of the plan creation device. Effect of the Invention

[0011] According to the present invention, it is possible to prevent the creation of a processing plan in which processing timings of objects of the same type are concentrated. [Brief description of the drawings]

[0012] [Figure 1] FIG. 2 illustrates an example of a functional configuration of a processing device. [Diagram 2] 11 is a flowchart illustrating an example of a processing method. [Diagram 3] FIG. 13 is a diagram illustrating an example of premise information when creating a loading port shipping plan. [Figure 4] FIG. 2 is a diagram illustrating an example of an outline of a loading port shipping plan. [Diagram 5] FIG. 2 is a diagram showing an example of a loading port shipping schedule for one loading port. [Figure 6] FIG. 13 is a diagram showing an example of a loading port shipping allocation plan for multiple loading ports created at one time. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that the fact that the length, position, size, interval, and other aspects of a comparison are the same includes not only cases where the comparison is strictly the same, but also cases where the comparison is different within the scope of the invention (for example, cases where the comparison is different within the range of tolerances determined at the time of design).

[0014] (Processing device and processing method) Fig. 1 is a diagram showing an example of a functional configuration of the processing device 110. Fig. 2 is a flowchart explaining an example of a processing method performed using the processing device 110. The processing device 110 creates a processing plan for processing objects. In this embodiment, a case is illustrated in which a processing plan is created that at least specifies the processing order of multiple objects. The processing plan may be, for example, a production plan or a logistics plan.

[0015] When the processing plan is a production plan, for example, the object is a product and the processing is various processing performed when manufacturing the product. When the processing plan is a logistics plan, for example, the object is a transportation object and the processing is various processing performed when transporting the transportation object.

[0016] The product is not limited to a finished product, but may be an intermediate product. The object of transportation may be a raw material, a material, an intermediate product, or a finished product. The means of transportation may be a ship, a vehicle such as a truck or a railroad car, or an airplane. As described above, the processing plan created by the processing device 110 is not limited.

[0017] As described above in the section on the problem to be solved by the invention, there are cases where it is required to process various objects in a time-distributed manner as much as possible. For example, a loading port ship allocation plan is an example of a processing plan that generates such a need. The loading port ship allocation plan includes, for example, a ship, a loading port to which the ship is allocated, the timing of the allocation of the ship to the loading port, and information on the attributes of the transport objects loaded on the ship. The transport objects are, for example, raw materials for producing (manufacturing) products such as steel products. In the loading port ship allocation plan, the timing of the allocation of the ship to the loading port corresponds to the timing of starting the transport of the raw materials loaded on the ship, and is an example of processing timing. In addition, "timing (processing timing)" means a point or period determined according to the unit of time in the processing plan (a division period described later), and in the case of a processing plan that is formulated on a daily basis, it may be a point at which processing is performed (transportation starts) somewhere on a determined day. For example, "timing (processing timing)" may be expressed by a date (year / month / day) and time, or may be expressed only by a date, or may be expressed only by a year and month.

[0018] For example, when producing pig iron in a blast furnace of a steelworks, multiple types of raw materials may be mixed in a predetermined ratio to stabilize the quality of the pig iron tapped from the blast furnace. In such a case, if the departure of the same type of raw materials from the loading port is biased to the same time, the same type of raw materials are likely to arrive at the steelworks at the same time, which may result in a shortage of some raw materials and the inability to mix and use multiple raw materials. In other words, if ships loaded with the same type of raw materials depart from the loading port at similar times, the same type of raw materials are more likely to arrive at the steelworks at similar times, and the more the arrival of the same type of raw materials is concentrated, the more likely it is that other types of raw materials that are needed will not arrive at the steelworks. Alternatively, if ships with the same brand of raw materials and similar loading capacities are dispatched at similar times, the amount of raw materials of the same brand that arrive at the steelworks will be unevenly distributed depending on time, which may cause an overstock or shortage of inventory. Therefore, the inventors focused on the loading patterns (types of raw materials and loading amounts) of the raw materials on ships departing from each loading port, and considered creating a loading port ship allocation plan so that the same loading patterns are distributed as much as possible in terms of time during the planning period.To achieve this, they considered it preferable to be able to create a loading port ship allocation plan taking into consideration not only the loading patterns of ships that are allocated immediately before and after each ship, but also the loading patterns of each ship and ships that are allocated significantly later than the allocation order of the ship in question.

[0019] Hereinafter, an example of the function of the processing device 110 of this embodiment and an example of the processing method of this embodiment will be described by taking an example of creating such a loading port shipping plan. In FIG. 1, the processing device 110 is an information processing device (computer) and includes, for example, one or more hardware processors and one or more memories as hardware. The processing device 110 executes various calculations by executing one or more programs stored in the memory by one or more hardware processors. The hardware processor may be, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The processing device 110 may include a CPU and a GPU. The memory may be, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory). The processing device 110 may include a RAM and a ROM. The processing device 110 may include a storage medium other than the RAM and the ROM. The processing device 110 may be realized by dedicated hardware such as, for example, an ASIC (Application Specific Integrated Circuit).

[0020] The input device 120 is a device for inputting various types of information to the processing device 110. For example, the input device 120 may include a database. Also, the input device 120 may include a user interface. In this case, the input device 120 includes, for example, a keyboard and a mouse. Also, the input device 120 may include an information processing device (computer) separate from the processing device 110. The number of input devices 120 may be one, or two or more. Also, at least one of the input devices 120 may be a device inside the processing device 110.

[0021] The output device 130 is a device that performs processing based on information output from the processing device 110. For example, the output device 130 may include a computer display. The output device 130 may also include an information processing device (computer) separate from the processing device 110. The output device 130 may also include a database that stores information output from the processing device 110. The number of output devices 130 may be one, or two or more. At least one of the output processing devices 130 may also be an internal device of the processing device 110. The communication between the input device 120 and the output device 130 and the processing device 110 may be wired communication or wireless communication.

[0022] In the present embodiment, a case in which the processing device 110 includes an acquisition unit 111, a planning period division unit 112, an evaluation index calculation unit 113, a processing plan creation unit 114, and an output unit 115 is illustrated in FIG.

[0023] <Acquisition unit 111, step S201> The acquisition unit 111 acquires at least premise information related to attributes of an object as premise information for creating a processing plan (step S201).

[0024] FIG. 3 is a diagram for explaining an example of premise information when creating a loading port shipping allocation plan. In the following description, the premise information when creating a loading port shipping allocation plan is abbreviated to premise information as necessary. In addition, in this embodiment, in addition to the information explicitly stated as premise information in this section, the acquisition unit 111 also acquires, as premise information, information that needs to be set in advance in the processing device 110 when creating a plan using a computer. For example, weight coefficients w1, w2 and a constant C of the objective function described later are acquired as premise information.

[0025] In FIG. 3, the ship ID is the identification information of the ship. In FIG. 3, the ship IDs (i1 to i6) of six ships are shown. In the following description, the ships with ship IDs i1, i2, i3, i4, i5, and i6 are referred to as ships i1, i2, i3, i4, i5, and i6 as necessary. The number of ships is not limited to six. In FIG. 3, A to C are the brands of raw materials. Different brands usually differ in any of the origin, components, characteristics (electrical, chemical, mechanical), and shape. In the following description, information that specifies the type of raw material is referred to as the brand as necessary. In the following description, raw materials with brands A, B, and C are referred to as raw materials A, B, and C as necessary. In FIG. 3, the height of the rectangle represents the loading amount of each of raw materials A to C. For example, in the ship i3, the loading amount of raw material A is larger than the loading amount of raw material C. In the present embodiment, the case where the loading amount is represented by mass (ton) is illustrated. In addition, the number of brands of raw materials is not limited to 3. In addition, Fig. 3 illustrates a case where the maximum loading capacity of each of the ships i1 to i6 is the same. However, the maximum loading capacity of each of the ships i1 to i6 may be the same or different.

[0026] In this embodiment, a case is exemplified in which the premise information includes a ship ID, information indicating the brand of raw materials, and information indicating the loading amount of raw materials for each ship and each brand. In this case, the brand and loading amount of raw materials are examples of attributes of the object. The brand of raw materials is an example of identification information of the type of raw materials, and the loading amount of raw materials is an example of the amount of raw materials. Therefore, in this embodiment, a case is exemplified in which the attribute of the object includes the type and amount of the object. However, since the attribute of the object is determined according to the object, it is not limited to these pieces of information. For example, ingredients may be included in the attribute of the object instead of or in addition to the brand. Also, at least one of the volume, size (e.g., length, width, height), and number may be included in the attribute of the object instead of or in addition to the amount. Also, other attributes (e.g., quality) may be included in the attribute of the object. Also, the attribute of the object included in the premise information may be of one type.

[0027] FIG. 4 is a diagram for explaining an example of the outline of the loading port ship allocation plan. In this embodiment, as illustrated with reference to FIG. 3, a case is illustrated in which the brand of raw materials loaded on each ship i1 to i6 and the loading amount of raw materials for each brand are predetermined as premise information. In addition, in this embodiment, a case is illustrated in which a loading port ship allocation plan is created by calculating the timing for allocating ships i1 to i6, whose brands and loading amounts for each brand are predetermined, to a loading port. The number of loading ports to which one ship is allocated may be one or may be two or more. When one ship (the same ship) is allocated to multiple loading ports, the timing for allocating the ship to each of the multiple loading ports is calculated. The timing for allocating a ship to a loading port corresponds to, for example, the timing when the ship departs from the loading port. Note that the timing for allocating a ship to a loading port may correspond to, for example, the timing when the ship calls at the loading port. In addition, in this embodiment, for the sake of simplicity, a case is illustrated in which the number of ships that can be allocated at the same time is one. However, the number of ships that can be assigned at the same time may be two or more. When the number of ships that can be assigned at the same time is two or more, the two or more ships may be assigned to the same loading port or to different loading ports. In the following explanation, the timing of assigning a ship to a loading port is referred to as the ship assignment timing as necessary. As described above, in this embodiment, the brands of raw materials to be loaded on each of the ships i1 to i6 and the loading amount of raw materials for each brand are determined in advance as premise information. Therefore, the order in which the ships are assigned corresponds to the order in which the raw materials are transported.

[0028] In FIG. 4, the loading port ID is identification information of the loading port. In FIG. 4, the loading port IDs (X1 to X2) of two loading ports are shown. In the following description, the loading ports with the loading port IDs X1 and X2 are referred to as loading ports X1 and X2 as necessary. Also, in FIG. 4, a case is illustrated in which raw material A or B is loaded onto a ship at the loading port X1, and raw material C is loaded onto a ship at the loading port X2. Note that the number of loading ports is not limited to two. The number of loading ports may be one, or may be three or more. Also, in FIG. 4, for the sake of simplicity, a case is illustrated in which only one of two brands of raw materials A and B is loaded onto one ship at the loading port X1. However, both of the two brands of raw materials A and B may be loaded onto one ship. Note that in this embodiment, raw materials loaded from different loading ports are considered to be different brands even if they have the same role in producing a product. In addition, in the loading port ship allocation plan shown in Figure 4, one bar graph extending in the vertical direction represents the loading state of raw materials on one ship. Furthermore, symbols A to C shown in the bar graph represent the brands of raw materials. In Figure 4, the loading volume of raw materials of the brand represented by the symbol is represented by the height of the rectangle in which the symbol is written.

[0029] In this embodiment, a case is illustrated in which the premise information further includes the planning period, the loading port ID, and the brand of the raw material to be loaded at each loading port. Note that, for example, if there is only one loading port, the loading port ID does not need to be included in the premise information.

[0030] In addition, the timing at which the acquisition unit 111 acquires each piece of premise information (ship ID, brand of raw material, amount of raw material loaded on each ship and for each brand, planning period, loading port ID, brand of raw material loaded at each loading port) may be different or the same.

[0031] In this embodiment, the input device 120 is illustrated as having a user interface. In this case, an operator operates the input device 120 to input premise information to the input device 120. However, as described above, the input device 120 is not limited to a device having a user interface.

[0032] <Planning period division unit 112, step S202> The planning period division unit 112 divides the planning period of the processing plan into a plurality of divided periods (step S202). In this embodiment, the planning period division unit 112 divides the planning period of the processing plan into a plurality of equal divided periods. However, the lengths of at least two of the plurality of divided periods may be different.

[0033] Here, it is preferable to be able to create a loading port ship allocation plan by considering not only the loading state of the raw materials loaded on the ships that are deployed immediately before or after each ship, but also the loading state of the raw materials loaded on each ship and on ships that are deployed significantly later than the shipping order of the ship. As described above, this is because a loading port ship allocation plan can be created so that the loading state of the raw materials (raw material brand and loading amount) on the ships departing from each loading port is dispersed as much as possible in terms of time, and as a result, it is expected that the brand and loading amount of the raw materials arriving at the discharge port will be dispersed in terms of time. Therefore, when creating a loading port ship allocation plan, it is preferable to consider the difference in the shipping timing of each ship (this will be explained again in the section <Processing plan creation unit 114, step S204> described later).

[0034] In this case, the difference in the ship allocation timing may be used as is. However, for example, in order to reduce the calculation load, this embodiment illustrates a case in which the difference in the ship allocation timing of each ship is expressed as the difference in the divided period to which the ship allocation timing belongs. In this case, the difference between multiple ship allocation timings belonging to the same divided period is considered to be zero (i.e., 0).

[0035] In this embodiment, the case where the planning period division unit 112 divides the planning target period into days is exemplified. When the planning target period is divided into days in this manner and the time axis unit of the loading port shipping plan is date (year, month, and day) (i.e., when the shipping timing is expressed in dates), the shipping timing and the divided period match. Also, for example, in order to further reduce the calculation load, the planning target period may be divided into periods longer than one day (for example, every five days). In this manner, the divided period is determined in advance by considering the trade-off between reducing the calculation load and increasing the accuracy of the processing plan, for example.

[0036] <Evaluation index calculation unit 113, step S203> The evaluation index calculation unit 113 calculates the value of an evaluation index for evaluating the processing plan using the premise information acquired by the acquisition unit 111 (step S203). In this embodiment, a case is illustrated in which the evaluation index calculation unit 113 calculates the value of an evaluation index for evaluating a loading port ship allocation plan. In addition, in this embodiment, a case is illustrated in which the evaluation index calculation unit 113 calculates the value of an attribute evaluation index that evaluates the degree of similarity of attributes of multiple objects. Specifically, in this embodiment, a case is illustrated in which the attribute evaluation index is an evaluation index that evaluates the degree of similarity of the stacking forms of raw materials loaded on multiple ships.

[0037] As described above, it is preferable to be able to create a loading port ship allocation plan by considering not only the loading state of raw materials loaded on ships that are allocated immediately before or after each ship, but also the loading state of raw materials loaded on each ship and on ships that are allocated significantly later than the allocation order of the ship. Therefore, in order to make this possible, in this embodiment, an example is given of a case in which the attribute evaluation index includes an evaluation index that evaluates the degree of similarity of attributes for all combinations of two objects that can be selected from the objects included in the processing plan.

[0038] As described above, the timing of ship allocation corresponds to the timing at which the transportation of raw materials begins. For example, evaluating the degree of similarity of attributes of raw materials loaded onto two ships with successive ship allocation orders corresponds to evaluating the similarity of attributes of raw materials with successive transportation orders. The transportation order (timing at which transportation begins) of raw materials loaded onto the same ship is the same.

[0039] The degree of similarity of attributes of a plurality of objects includes, for example, at least one of the similarity (degree of similarity) of attributes of a plurality of objects and the dissimilarity (degree of dissimilarity) of attributes of a plurality of objects. For example, the degree of similarity between attributes of a plurality of objects may be expressed by two values ​​(for example, "0" and "1") according to the presence or absence of a difference between the (similar) attributes of the plurality of objects. Also, for example, the degree of similarity of attributes of a plurality of objects may be expressed by a value corresponding to the difference in values ​​indicating (the same type of) attributes of the plurality of objects. Also, for example, the degree of similarity of attributes of a plurality of objects may be expressed as a value according to the number or ratio of common (similar) attributes among the plurality of objects. Also, for example, the degree of similarity of attributes of a plurality of objects may be expressed as a value according to the number or ratio of different (similar) attributes among the plurality of objects.

[0040] As described above, in this embodiment, a case is exemplified in which the premise information includes the raw material brand and loading amount as raw material attributes. Therefore, in this embodiment, a case is exemplified in which the evaluation index calculation unit 113 calculates an attribute evaluation index for evaluating the degree of similarity of the raw material brands and an attribute evaluation index for evaluating the degree of similarity of the raw material loading amount. In the following description, the attribute evaluation index for evaluating the degree of similarity of the raw material brands is referred to as a brand difference evaluation index as necessary. Also, the attribute evaluation index for evaluating the degree of similarity of the raw material loading amount is referred to as a loading amount difference evaluation index as necessary.

[0041] Here, if the processing order relationship is calculated for each of two objects selectable from the objects included in the processing plan, the processing order of the objects can be calculated without a high calculation load compared to the case of calculating the processing order relationship for each of three or more objects. In this case, the processing order of all objects included in the processing plan can be calculated by calculating the processing order relationship for each of all two objects selectable from the processing plan. Therefore, in this embodiment, a case is illustrated in which the evaluation index calculation unit 113 calculates the attribute evaluation index for each of all two ships selectable from the ships included in the loading port shipping plan. However, this is not necessarily required. For example, if there is a ship that needs to be shipped at a specific time among the ships included in the loading port shipping plan, the shipping timing of the ship may be fixed. In this case, the raw materials loaded on the ship may or may not be included in the calculation target of the attribute evaluation index. In this case, a constraint may be imposed to fix the timing of dispatch of the ship in question, and then the dispatch order (more specifically, the timing of dispatch) of other ships may be calculated (this will be explained again in the section <Processing plan creation unit 114, step S204> described later).

[0042] Specifically, in this embodiment, a case will be illustrated in which the evaluation index calculation unit 113 calculates the following attribute evaluation indexes (brand difference evaluation index, loading amount difference evaluation index).

[0043] From the viewpoint of preventing ships with similar shipping timing from loading the same brand of raw materials on two ships, this embodiment illustrates a case where the brand difference evaluation index is expressed using the number of brands of raw materials common to the two ships among the brands of raw materials loaded on the two ships. For example, in FIG. 3, the brands of raw materials loaded on ships i1 and i2 are both B and C. Therefore, the value of the brand difference evaluation index for ships i1 and i2 is 2. Moreover, of the brands of raw materials loaded on ships i1 and i3, the only brand common to ships i1 and i3 is C. Therefore, the brand difference evaluation index for ships i1 and i3 is 1. However, this embodiment illustrates a case where the brand difference evaluation index is weighted based on the number of brands common to the two ships among the brands of raw materials loaded on such two ships, as will be described later.

[0044] In this manner, in this embodiment, a case is illustrated in which the greater the number of common brands of raw materials loaded onto two ships, the greater the value of the brand difference evaluation index for the two ships. In this case, a greater value of the brand difference evaluation index for the two ships corresponds to a higher similarity in the attributes of the raw materials loaded onto the two ships.

[0045] For example, the brand difference evaluation index for two ships may be set so that the greater the number of brands of raw materials not in common among the brands of raw materials loaded on the two ships, the larger the brand difference evaluation index becomes. For example, in Fig. 3, of the brands of raw materials loaded on ships i2 and i3, the brands that are not in common between ships i2 and i3 are A and B. In this case, the value of the brand difference evaluation index is 2. In this case, a large value of the brand difference evaluation index for two ships corresponds to a low similarity in the attributes of the raw materials loaded on the two ships.

[0046] In addition, from the viewpoint of increasing the difference in loading amount between ships with close shipping timing, this embodiment illustrates a case where the loading amount difference evaluation index is expressed using the reciprocal of the sum of the absolute value of the difference in loading amount for each brand of raw material loaded on two ships. For example, in FIG. 3, when the loading amounts of brands B and C loaded on ship i1 are 50,000 tons and 30,000 tons, respectively, and the loading amounts of brands B and C loaded on ship i2 are 30,000 tons and 50,000 tons, respectively, the value of the loading amount difference evaluation index on ships i1 and i2 is, for example, 1 / 4 (=1 / {|5-3|+|3-5|}). In addition, in FIG. 3, when the loading amounts of brands A and C loaded on ship i3 are 60,000 tons and 20,000 tons, respectively, the value of the loading amount difference evaluation index on ships i2 and i3 is, for example, 1 / 12 (=1 / {|0-6|+|3-0|+|5-2|}). However, in this embodiment, as described below, an example is given in which the loading volume evaluation index is calculated by weighting the inverse sum of the absolute value of the difference in loading volume for each brand of raw material loaded onto two such ships.

[0047] In this manner, in this embodiment, a case is illustrated in which the greater the difference in loading volume for each brand of raw materials loaded onto two ships, the smaller the value of the loading volume difference evaluation index for the two ships. In this case, a smaller value of the brand difference evaluation index for the two ships corresponds to a lower similarity in the attributes of the raw materials loaded onto the two ships (in other words, a larger value of the brand difference evaluation index for the two ships corresponds to a higher similarity in the attributes of the raw materials loaded onto the two ships).

[0048] For example, the loading capacity difference evaluation index for two ships may be determined so that the smaller the difference in the loading capacity of each brand of raw materials loaded on the two ships, the smaller the loading capacity difference evaluation index for the two ships. For example, the loading capacity difference evaluation index for two ships may be expressed as the sum of the absolute values ​​of the difference in the loading capacity of each brand of raw materials loaded on the two ships. In this case, a small value of the brand difference evaluation index for two ships corresponds to a high similarity in the attributes of the raw materials loaded on the two ships.

[0049] In this embodiment, an example is shown in which the evaluation index calculation unit 113 calculates the above brand difference evaluation index and load capacity difference evaluation index for each of all two ships that can be selected from the ships that can be included in the loading port shipping plan.

[0050] <Processing plan creation unit 114, step S204> The processing plan creation unit 114 creates a processing plan based on the evaluation result of the processing plan using the values ​​of the evaluation indexes calculated by the evaluation index calculation unit 113 (step S204). In this embodiment, the processing plan creation unit 114 creates a loading port ship allocation plan based on the evaluation result of the loading port ship allocation plan using the values ​​of the attribute evaluation indexes (brand difference evaluation index and loading capacity difference evaluation index).

[0051] Here, for example, from the viewpoint of dispersing the objects (in this embodiment, the stacking state of the raw materials on each ship), when the difference in the processing timing of two objects is small, it may be more important to make the attributes of the two objects non-identical and non-similar than to make the attributes of the two objects non-identical and non-similar when the difference in the processing timing of the two objects is large. Therefore, as described above in <Planning period division unit 112, step S202>, in order to take into account the difference in the dispatch timing of each ship when creating a loading port dispatch plan, this embodiment illustrates a case in which weighting is performed on the attribute evaluation index that evaluates the degree of similarity of the attributes of two objects based on the difference in the processing timing of the two objects (the difference in the dispatch timing in this embodiment). In the following description, the difference in processing timing is referred to as the processing time difference as necessary.

[0052] For example, the weight (weight coefficient) may be made to change monotonically with respect to the change in the processing time difference, so that the smaller the processing time difference between two objects, the larger the weight (weight coefficient). However, this is not necessarily required. For example, the range expected as the processing time difference may be divided into a plurality of ranges, and a weight (weight coefficient) may be determined in advance for each of the plurality of ranges. In this case, even if the processing time differences are different, if these processing time differences belong to the same range, the weight (weight coefficient) corresponding to these processing time differences will be the same.

[0053] It is not necessary to weight the two objects based on the processing time difference (difference in ship dispatch timing in this embodiment). For example, the attribute evaluation index for evaluating the degree of similarity of the attributes of the two objects may be weighted to balance the evaluation with other attribute evaluation indexes instead of or in addition to weighting based on the processing time difference (difference in ship dispatch timing in this embodiment) of the two objects (see weight coefficients w1 and w2 described later). It is not necessary to evaluate the processing plan using the processing time difference of the two objects. For example, a constraint may be imposed that the integrated value of the attribute evaluation index for N objects (N is an integer of 2 or more) whose processing order is consecutive is equal to or less than a predetermined value. It is also possible to impose such a constraint and to weight the N objects whose processing order is consecutive, for each attribute evaluation index, so that the sum of the integrated values ​​is equal to or less than a predetermined value. In addition, both imposing such a constraint and weighting based on the processing time difference of the two objects described above may be performed.

[0054] In the following description, the weighting based on the processing time difference between two objects is referred to as the weighting based on the processing time difference as necessary. Also, the weighting coefficient multiplied by the attribute evaluation index to perform the weighting based on the processing time difference is referred to as the weighting coefficient based on the processing time difference as necessary.

[0055] As described above, in this embodiment, the processing time difference between two objects is represented by the difference between the dispatch timings of two ships, and the difference between the dispatch timings of two ships is represented by the difference between the divided periods to which the dispatch timings of the two ships belong. For example, if the divided period to which the dispatch timing of ship i1 belongs is August 1, 2023, and the divided period to which the dispatch timing of ship i2 belongs is August 3, 2023, the difference between the divided periods to which the dispatch timings of ships i1 and i2 belong is 2 days (=3-1). In this embodiment, the difference between the divided periods is represented by the smallest element among the elements that make up the unit of the divided period (for example, if the unit of the divided period is year, month, and day, the smallest element among the elements that make up the unit, year, month, and day, is day, so the unit of the difference between the divided periods is day).

[0056] From the above viewpoint, in this embodiment, a case where the processing plan creation unit 114 evaluates the loading port ship allocation plan using the following formulas (1) and (2) is illustrated. Also, in this embodiment, the processing plan creation unit 114 evaluates the loading port ship allocation plan using the objective function J k We will explain the case where a loading port ship allocation plan is created by solving the minimization problem that minimizes the value of the objective function J k is an example of an evaluation result of the processing plan (in this embodiment, the loading port ship allocation plan). The optimization problem is not limited to a minimization problem, but may be a maximization problem. In this case, for example, the entire right side of equation (1) multiplied by (-1) may be used as the objective function. The optimization problem may be, for example, a mathematical optimization problem or a combinatorial optimization problem. The solution method of the optimization problem is not limited, and any known solution method may be used. For example, the solution method of the optimization problem may be a strict method using a branch and bound method that is generally used as a solution method for mathematical optimization problems, or may be an empirical method such as metaheuristics (for example, a genetic algorithm).

[0057]

number

[0058] Here, i and j indicate the ship ID. In the following explanation, ships with ship IDs i and j will be referred to as ships i and j as necessary. s and u indicate the timing of ship allocation (any divided period within the planning period). k indicates the loading port ID. In the following explanation, a loading port (loading port) with loading port ID k will be referred to as loading port k as necessary. There are as many formulas (1) as there are loading ports k. Note that if there is only one loading port k, k in formulas (1) and (2) is optional.

[0059] Σ i,s,j,u indicates that the sum is calculated for each of i, s, j, and u. δ i,s,k is a variable that is 1 when ship i is dispatched to loading port k at dispatch timing s, and is 0 when ship i is not dispatched to loading port k at dispatch timing s. j,u,k is a variable that is 1 when ship j is dispatched to loading port k at dispatch timing u, and is 0 when ship i is not dispatched to loading port k at dispatch timing s. i,s,k , δ j,u,k are the decision variables (design variables) sought in the optimization problem.

[0060] cost i,s,j,u,k is the case where ship i is dispatched to port k at dispatch timing s and ship j is dispatched to port k at dispatch timing u (i.e., decision variable δ i,s,k , δ j,u,k In this embodiment, the optimization problem is a minimization problem. i,s,j,u,k The smaller the value of , the better the loading port allocation plan in which ship i is allocated to loading port k at allocation timing s and ship j is allocated to loading port k at allocation timing u is evaluated as a good plan.

[0061] w1·name(i,j,k) / {time(i,j,k) 2 +C} is an example of a brand difference evaluation index between ship i that is to be dispatched to loading port k at dispatch timing s and ship j that is to be dispatched to loading port k at dispatch timing u. w2 / load(i,j,k) / {time(i,j,k) 2+C} is an example of a load capacity difference evaluation index between ship i that is assigned to loading port k at ship assignment timing s and ship j that is assigned to loading port k at ship assignment timing u.

[0062] name(i,j,k) is the number of brands of raw materials common to two ships i and j loaded at loading port k. 1 / load(i,j,k) is the reciprocal of the sum of the absolute values ​​of the differences in the loading volumes of each brand of raw material loaded onto two ships i and j at loading port k. Furthermore, w1 is a weighting coefficient by which name(i,j,k) is multiplied to weight name(i,j,k). w2 is a weighting coefficient by which 1 / load(i,j,k) is multiplied to weight 1 / load(i,j,k). In the above example, name(i,j,k) has no unit (it is a dimensionless quantity). Furthermore, the unit of 1 / load(i,j,k) is, for example, ton. Therefore, the values ​​of each evaluation index cannot be simply compared. Therefore, the weighting coefficients w1 and w2 can be used to balance the evaluation of each evaluation index.

[0063] Furthermore, the weighting coefficients w1 and w2 have values ​​that reflect the influence of each evaluation index. For example, when the weighting coefficients w1 and w2 are positive values, a large value of the weighting coefficient w1 indicates that importance is placed on name(i,j,k). On the other hand, a large value of the weighting coefficient w2 indicates that importance is placed on 1 / load(i,j,k). Note that one of the weighting coefficients w1 and w2 may be set to 1. Furthermore, it is preferable to include the weighting coefficients w1 and w2 in the objective function J, but it is also possible to include the weighting coefficients w1 and w2 in the objective function J. k (In other words, all weighting coefficients w1 and w2 may be 1).

[0064] 1 / {time(i,j,k) 2+C} is a weighting coefficient calculated based on the difference in the division period between ship i that is dispatched to loading port k at dispatch timing s and ship j that is dispatched to loading port k at dispatch timing u. In this embodiment, 1 / {time(i,j,k) 2 +C} is weighted by (1 / {time(i,j,k) 2 +C} is multiplied as a weighting coefficient. time(i,j,k) is the absolute value of the difference in the divided periods between ship i that is to be dispatched to loading port k at dispatch timing s and ship j that is to be dispatched to loading port k at dispatch timing u. In the following explanation, time(i,j,k) will be referred to as the divided period difference time(i,j,k) as necessary. C is the cost difference when the divided period difference time(i,j,k) is 0. i,s,j,u,k is a positive constant to prevent from becoming infinity.

[0065] In this embodiment, as shown in equation (2), a case where name(i,j,k), 1 / load(i,j,k), and time(i,j,k) are each squared is illustrated. k (cost i,s,j,u,k However, if we do so, the objective function J may vary depending on whether the difference between the evaluation indexes for two ships i and j and the evaluation indexes for two other ships i and j is large or small. k (cost i,s,j,u,k ) may not change (or may become close to it).

[0066] For example, if the value of an evaluation index for two ships i and j is 1, and the value of an evaluation index for two other ships i and j is 4, the absolute value of the difference between those two evaluation indexes is 3 (=4-1). Also, if the value of an evaluation index for two ships i and j is 2, and the value of an evaluation index for two other ships i and j is 3, the absolute value of the difference between those two evaluation indexes is 1 (=3-2). Therefore, the objective function J calculated using the two evaluation indexes when the former two evaluation indexes are used and when the latter two evaluation indexes are used arek However, the sum (integrated value) of the former two evaluation indexes and the sum (integrated value) of the latter two evaluation indexes are both 5. Therefore, it is not necessary to square the evaluation indexes as in equation (2) to calculate the cost i,s,j,u,k Calculate each cost without squaring the evaluation index. i,s,j,u,k When the objective function J is calculated as in equation (1), the objective function J k The value of and the objective function J when the latter two evaluation indices are used. k The value of and does not change (or becomes close), and the objective function J k In other words, even though the evaluation index values ​​are different, the quality of the loading port and shipping allocation plans of both companies will be easily evaluated as being of the same level.

[0067] On the other hand, if we square the evaluation index, the objective function J k For example, in the above example, the sum of the squares of the first two evaluation indices is 17 (=4 2 +1 2 ) On the other hand, the sum of the squares of the latter two evaluation indices is 13 (=3 2 +2 2 ) Therefore, rather than using the values ​​of the evaluation indices as they are, the objective function J k The value of and the objective function J when the latter two evaluation indices are used. k Therefore, for example, when creating a loading port vessel allocation plan that aims to calculate an optimal solution that reduces the difference between multiple evaluation indexes of the same type, it is preferable to square the evaluation index. The above (that squaring is preferable when calculating the value of the objective function) also applies to the processing time difference. Note that it is also possible to square the values ​​of some types of evaluation indexes without squaring the values ​​of all types of evaluation indexes. In this case, the processing time difference may or may not be squared. Also, it is also possible to square the processing time difference without squaring the values ​​of all types of evaluation indexes. In other words, it is also possible to square at least one of some or all types of evaluation indexes and the processing time difference.

[0068] In addition, two values ​​with the same absolute value but different signs will have the same value when squared. Therefore, by squaring the evaluation index and the processing time difference, it is not necessary to calculate the absolute value when calculating the difference.

[0069] In equation (2), the smaller the division period difference time(i,j,k) (processing time difference) between the dispatch timings of two ships i and j, the smaller the cost cost i,s,j,u,k Therefore, in equation (2), the smaller the division period difference time(i,j,k) (processing time difference) between the dispatch timings s and u of two ships i and j, the smaller the objective function J k Therefore, in equation (2), the smaller the difference (processing time difference) between the division periods to which the dispatch timings s and u of two ships i and j belong, the more likely it is that the evaluation result of the loading port dispatch plan will be judged to be poor (i.e., it will be more difficult to adopt it as the optimal solution).

[0070] The optimization problems shown in the above equations (1) and (2) are k The value of the variable δ when considered to be the smallest in the optimization problem algorithm while satisfying the constraints. i,s,k , δ j,u,k This is a question that asks:

[0071] The constraints may be determined so that the loading port ship allocation plan can be calculated appropriately. For example, the constraints may include that all ships included in the loading port ship allocation plan are allocated once to the loading port k where the raw materials loaded on the ship are located. The constraints may also include that a predetermined number or more of ships among the ships i included in the loading port ship allocation plan are allocated once to the loading port where the raw materials loaded on the ship are located. The constraints may also include that at least one ship among all the ships included in the loading port ship allocation plan cannot be allocated at a specific ship allocation timing. At least one part of the ships that are the subject of the loading port ship allocation plan, the loading ports included in the loading port ship allocation plan as the loading ports of the ships, and the ship allocation timing may be excluded from the formulas (1) and (2). In this case, the calculations of the formulas (1) and (2) may be performed by treating the ones that are excluded from the formulas (1) and (2) among i, j, s, u, and k as missing numbers (as if they do not exist).

[0072] In addition, as described above in the section <Evaluation index calculation unit 113, step S203>, when there is a ship that needs to be dispatched at a specific time, the dispatch timing of the ship may be included in the constraint conditions. A constraint equation, which is a mathematical expression expressing such a constraint condition, may be formulated according to a solution method for the optimization problem.

[0073] The processing schedule creation unit 114 calculates the decision variable δ using the above equations (1) and (2). i,s,k , δ j,u,k In this way, for example, among the ships included in the loading port shipping schedule, the shipping timings of all ships to be shipped to one loading port k are calculated. The processing plan creation unit 114 calculates the optimal solution of such a decision variable δ i,s,k , δ j,u,k The optimal solution for each loading port k is calculated. This allows the loading port vessel allocation plan for each loading port k to be calculated in sequence.

[0074] For example, the processing plan creation unit 114 may set a decision variable δ i,s,k , δ j,u,kBased on the optimal solution, information indicating the ship ID of each ship and the dispatch timing of the ship with the ship ID for each loading port k is calculated as processing plan information indicating a loading port ship dispatch plan. When solving the optimization problem, for example, the search for a solution is repeated until a convergence condition is satisfied. In this case, the processing plan (goodness or badness) is evaluated by determining whether the convergence condition is satisfied or not.

[0075] As described above in the section <Evaluation index calculation unit 113, step S203>, for example, the value of the brand difference evaluation index may be determined so that the greater the number of brands of raw materials not in common loaded onto the two ships, the greater the value of the brand difference evaluation index. In this case, for example, the weighting coefficient w1 may be multiplied by the inverse of the brand difference evaluation index (instead of the brand difference evaluation index). Also, the value of the loading capacity difference evaluation index for the two ships may be indicated as being smaller so that the smaller the difference in loading capacity between the brands of raw materials loaded onto the two ships, the smaller the value of the loading capacity difference evaluation index. In this case, for example, the weighting coefficient w2 may be multiplied by the loading capacity difference evaluation index (instead of the inverse of the loading capacity difference evaluation index).

[0076] As mentioned above, weighting based on the processing time difference is not necessarily required. For example, a constraint may be imposed that the integrated value of the attribute evaluation index for N objects (N is an integer of 2 or more) that are processed consecutively must be less than a predetermined value. In this case, the denominator (time(i,j,k)) on the right side of equation (2) 2 +C) is 1. For example, for N consecutive (N is an integer of 2 or more) delivery times s and u for ships i and j at a certain loading port k, "Σ i,s,j,u name(i,j,k) 2 +1 / load(i,j,k) 2 It is also possible to include in the constraint that "≦TH" is satisfied. In this way, it is not necessary to (explicitly) calculate the processing time difference. However, such a constraint may be imposed without modifying formula (2).

[0077] Furthermore, if an optimal solution is calculated by solving an optimal solution problem, a better processing plan can be created in a short time, which is preferable. However, it is not always necessary to solve an optimal solution problem. For example, the allocation timings s and j of ships i and j that satisfy predetermined conditions may be searched for by trial and error based on the brand difference evaluation index and the load capacity difference evaluation index. This may be done when the number of ships i and j and the brands of raw materials is small. In this case, the processing plan (is good or bad) is evaluated by judging whether the predetermined conditions are satisfied or not.

[0078] <Output Unit 115, Step S205> The output unit 115 outputs the processing plan information (step S205). In this embodiment, a case where the output unit 115 outputs the processing plan information to the output device 130 is illustrated. In addition, when the output device 130 includes a computer display, the processing plan information output from the output unit 115 is displayed on the computer display. As described above, the output device 130 is not limited to a device including a computer display.

[0079] FIG. 5 is a diagram showing an example of a loading port ship allocation plan for one loading port created by the method of this embodiment. In FIG. 5, n+5, n+10, . . . , and n+30 on the horizontal axis indicate 5 days, 10 days, . . . , and 30 days after the first date (year / month / day) n of the planning period, respectively. The loading volume on the vertical axis is a standardized volume (dimensionless volume). The method of notating the bar graph in FIG. 5 is the same as that of the bar graph in FIG. 4. Note that FIG. 4 illustrates an example in which only one brand of raw material is loaded on one ship. FIG. 5 illustrates an example in which multiple brands of raw materials may be loaded on one ship (for example, see bar graphs 501 to 503 showing the loading volume of a ship on which raw materials of brands D and E are loaded). As shown in FIG. 5, by using the method of this embodiment, a loading port ship allocation plan is created so that ships with similar loading patterns are not placed in succession, and the loading patterns are dispersed (see, for example, bar graphs 501 to 503).

[0080] (summary) As described above, in this embodiment, the processing device 110 uses premise information regarding the attributes of the objects to calculate the value of an attribute evaluation index that evaluates the degree of similarity of the attributes of multiple objects, and creates a processing plan based on the result of evaluating the processing plan using the value of the attribute evaluation index. At this time, as the degree of similarity of the attributes of the multiple objects evaluated by the attribute evaluation index, the degree of similarity of the attributes of at least some combinations of two objects selectable from the objects included in the processing plan is used. Therefore, for example, the processing order of each object can be determined by considering not only the objects processed immediately before or after each object, but also the objects processed significantly later than the processing order of each object and the object. Therefore, it is possible to prevent the creation of a processing plan in which the processing timings of the same type of objects are concentrated. Here, as the degree of similarity of the attributes of the multiple objects evaluated by the attribute evaluation index, the degree of similarity of all combinations of two objects selectable from the objects included in the processing plan may be used, or the degree of similarity of some combinations of all combinations of two objects selectable from the objects included in the processing plan may be used. For example, if there is an object in the processing plan that has no restrictions on the processing order, the object can be excluded from the objects to be included in the processing plan, and the value of the attribute evaluation index that evaluates the degree of similarity of the attributes for a combination of two selectable objects can be calculated.

[0081] In addition, in this embodiment, the processing device 110 includes a weighting coefficient based on the processing time difference between two objects in the attribute evaluation index that evaluates the degree of similarity of the attributes of the two objects. Therefore, even if the attributes of the two objects are the same, a processing plan can be created that takes into account the processing time difference between the two objects. Therefore, a processing plan that is more suited to the above-mentioned needs can be created.

[0082] In this embodiment, the processing device 110 includes a weighting coefficient for weighting the squared processing time difference between two objects in the attribute evaluation index that evaluates the degree of similarity of the attributes of two objects. Therefore, when evaluating the processing plan (the quality of the processing plan), the difference in the processing time difference can be clearly expressed.

[0083] In addition, in this embodiment, the processing device 110 divides the planning period for the processing plan into a plurality of divided periods, and the difference in processing timing (processing time difference) between two objects is set as the difference between the divided periods to which the processing timing belongs. Therefore, for example, it is possible to reduce the calculation load when creating a processing plan compared to calculating the difference in processing timing between the two objects by directly using the processing timing of the objects.

[0084] In addition, in this embodiment, the processing device 110 calculates the value of an attribute evaluation index including an attribute evaluation index for evaluating the degree of similarity of the types of objects and an attribute evaluation index for evaluating the degree of similarity of the quantities of the objects, and creates a processing plan based on the result of evaluating the processing plan using the attribute evaluation index value. Therefore, for example, for objects of many types and small lots (small amounts of each type), it is possible to prevent the creation of a processing plan in which the processing timings of objects of the same type are concentrated.

[0085] Furthermore, in this embodiment, the processing device 110 creates a plan for one or more locations as a processing plan. Therefore, a processing plan for each location can be created. In the example described in this embodiment, the loading port corresponds to the location. Furthermore, for example, when the processing plan is a production plan, for example, a production line or a factory corresponds to the location. Furthermore, for example, when creating a production plan for manufacturing a product to be processed on multiple production lines or a production plan for manufacturing a product to be processed on multiple factories, a production plan for multiple production lines or multiple factories may be created.

[0086] In addition, in this embodiment, the processing device 110 creates a processing plan using a value obtained by squaring the value of at least one type of attribute evaluation index. Therefore, when evaluating the processing plan (whether it is good or bad), the difference between the same type of attribute evaluation indexes can be made prominent.

[0087] In this embodiment, the processing device 110 creates a processing plan by solving an optimization problem in which a mathematical expression having an attribute evaluation index is used as an objective function. Therefore, the processing plan can be calculated quickly and with high accuracy.

[0088] (Modification) In this embodiment, the processing plan creation unit 114 sequentially calculates the loading port shipping allocation plan for each loading port k as the loading port shipping allocation plan. When creating loading port shipping allocation plans for multiple loading ports k, it may be required to create a loading port shipping allocation plan not only for each loading port k but also when the loading port shipping allocation plans for multiple loading ports k are viewed collectively, so that ships with similar loading configurations as much as possible are not allocated at similar times. Therefore, the processing plan creation unit 114 may create loading port shipping allocation plans for multiple loading ports k at once using the values ​​of the attribute evaluation indexes for multiple loading ports k.

[0089] In this case, for example, the objective function J in Eq. (1) k Instead of the above, the objective function J in the following equation (3) may be used. i,s,j,u,k is expressed by equation (2).

[0090]

number

[0091] Here, Σ i,s,j,u,k indicates that the integration is performed for each of i, s, j, u, and k. k is the weighting coefficient for the product k. k By using this, the degree of temporal variation in the loading status of a ship can be adjusted for each loading port k. In the example shown in equations (2) and (3), when the degree of temporal variation in the loading status of a ship at a certain loading port k is increased, the weighting coefficient ω k On the other hand, in another loading port k, if the degree of variation in the loading status of the vessel over time does not need to be increased so much, the weighting coefficient ω k In this way, the weighting coefficient ω kIt is preferable to include in the objective function J, but the weighting factor ω k does not need to be included in the objective function J (in other words, all weight coefficients ω k can be 1).

[0092] The above-mentioned formula (1) exists for the number of loading points k. On the other hand, the number of formulas (3) is 1 regardless of the number of loading points k. The processing schedule creation unit 114 uses formulas (2) and (3) to calculate the decision variable δ i,s,k , δ j,u,k By calculating the optimal solution of i,s,k , δ j,u,k When calculating the optimal solution (objective function J), the attribute evaluation index (w1·name(i,j,k / {time(i,j,k) 2 +C}, w2 / load(i,j,k) / {time(i,j,k) 2 In addition, the decision variables δ for multiple shipping ports k are calculated by using equation (3). i,s,k , δ j,u,k The optimal solutions for are calculated simultaneously. Therefore, the processing plan creation unit 114 can create loading port ship allocation plans for multiple loading ports k at once.

[0093] FIG. 6 is a diagram showing an example of a loading port ship allocation plan for multiple loading ports k created at once as described in this modified example. The vertical axis, horizontal axis, and bar graph notation methods in FIG. 6 are the same as those in FIG. 5. FIG. 6(a) is a diagram showing an example of a loading port ship allocation plan for loading port X1. FIG. 6(b) is a diagram showing an example of a loading port ship allocation plan for loading port X2. FIG. 6(c) is a diagram showing the loading port ship allocation plans for loading ports X1 and X2 together. Note that loading ports X1 and X2 correspond to loading ports X1 and X2 shown in FIG. 4, but the loading forms of each ship are different from the loading forms exemplified in FIG. 3. As shown in FIG. 6(a) to FIG. 6(c), by using the method of this modified example, it can be seen that the loading forms of each ship are dispersed over time not only for loading ports X1 and X2 but also for loading ports X1 and X2 together.

[0094] Furthermore, in this embodiment, an example has been given of a case in which raw materials to be loaded onto a ship are determined for each raw material of the same brand. However, this does not necessarily have to be the case. For example, raw materials may be grouped based on premise information regarding the attributes of the raw materials. For example, raw materials of multiple brands that have similar raw material quality (values ​​indicating quality within a predetermined range) may be placed in the same group. In this case, raw materials to be loaded onto a ship are determined for each group. Furthermore, the attribute evaluation index is an evaluation index that evaluates the difference in the attributes of raw materials for each group.

[0095] When determining an attribute evaluation index in this manner, for example, in the description of this embodiment, the attribute of the raw material may be replaced with the attribute of the group. In this case, one group may contain multiple values ​​for the same type of attribute. In that case, for example, a representative value (such as the average or median) of multiple values ​​of the same type of attribute contained in one group may be set as the value of the attribute of the group. In addition, for example, the brand may be quantified. For example, the brand may be quantified so that the closer the quality of the brands, the closer the numerical value.

[0096] In addition, the cost calculated using the attributes of the raw materials as described in this embodiment i,s,j,u,k and the cost calculated using the group attributes as in this modified example. i,s,j,u,k In this case, the objective function J k For example, the former cost i,s,j,u,k The integrated value Σ i,s,j,u cost i,s,j,u,k And the latter cost i,s,j,u,k The integrated value Σ i,s,j,u cost i,s,j,u,k It is expressed as a weighted linear sum of and.

[0097] That is, the former cost i,s,j,u,k cost1 i,s,j,u,k The latter cost is i,s,j,u,k cost1 i,s,j,u,k Then, the objective function J k For example, "J k =w3 Σ i,s,j,ucost1 i,s,j,u,k δ i,s,k δ j,u,k +w3·Σ i,s,j,u cost2 i,s,j,u,k δ i,s,k δ j,u,k " w3 and w4 are respectively expressed as cost1 i,s,j,u,k , cost2 i,s,j,u,k is the weighting factor for

[0098] (Other embodiments) The above-described embodiment of the present invention can be realized by a computer executing a program. A computer-readable recording medium having the program recorded thereon and a computer program product such as the program can also be applied as an embodiment of the present invention. Examples of the recording medium that can be used include a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a non-volatile memory card, and a ROM. The embodiment of the present invention can be realized by a PLC (Programmable Logic Controller) or dedicated hardware such as an ASIC (Application Specific Integrated Circuit). Furthermore, the above-described embodiments of the present invention are merely examples of the implementation of the present invention, and the technical scope of the present invention should not be interpreted as being limited by these. In other words, the present invention can be implemented in various forms without departing from its technical concept or main features.

[0099] The disclosure of the above embodiment can be, for example, as follows. [Disclosure 1] A planning device for creating a processing plan for processing a plurality of objects, comprising: an acquisition unit that acquires at least premise information regarding attributes of the plurality of objects as premise information for creating the processing plan; an evaluation index calculation unit that calculates an evaluation index value for evaluating the processing plan by using the premise information; a processing plan creation unit that creates the processing plan based on an evaluation result of the processing plan using the value of the evaluation index; Equipped with the evaluation index includes an attribute evaluation index for evaluating the degree of similarity of attributes of the plurality of objects; A plan creation device, wherein the degree of similarity of the attributes of the multiple objects evaluated using the attribute evaluation index includes the degree of similarity of attributes for at least some combinations of two objects selectable from the objects to be included in the processing plan. [Disclosure 2] The plan creation device described in Disclosure 1, wherein the degree of similarity of the attributes of the multiple objects evaluated by the attribute evaluation index includes the degree of similarity of attributes for all combinations of two objects selectable from the objects to be included in the processing plan. [Disclosure 3] The plan creation device according to Disclosure 1 or 2, wherein the evaluation index includes a weighting coefficient for the attribute evaluation index that is based on a difference in processing timing between two objects among the plurality of objects. [Disclosure 4] The plan creation device according to disclosure 3, wherein the weighting coefficient is a coefficient for weighting a value obtained by squaring the difference. [Disclosure 5] A planning period division unit that divides a planning target period of the processing plan into a plurality of divided periods, The plan creation device according to Disclosure 3 or 4, wherein the difference in the processing timing is represented by the difference in the divided periods to which the processing timing belongs. [Disclosure 6] The plan creation device according to any one of Disclosures 1 to 5, wherein the attribute evaluation index includes an attribute evaluation index that evaluates the degree of similarity of the type of the object, and an attribute evaluation index that evaluates the degree of similarity of the quantity of the object. [Disclosure 7] The plan creation device according to any one of Disclosures 1 to 6, wherein the processing plan includes plans at one or more locations. [Disclosure 8] The attribute evaluation index is an evaluation index for each location, The plan creation device according to Disclosure 7, wherein the processing plan creation unit creates the processing plans for the multiple locations at once by using values ​​of the attribute evaluation indexes for the multiple locations together. [Disclosure 9] The evaluation index calculation unit groups the objects based on the premise information, The plan creating device according to any one of Disclosures 1 to 8, wherein the attribute evaluation index is an evaluation index for evaluating a degree of similarity of attributes of the objects for each of a plurality of groups. [Disclosure 10] The plan creation device according to any one of Disclosures 1 to 9, wherein the processing plan creation unit creates the processing plan using a value obtained by squaring a value of at least one type of the attribute evaluation index. [Disclosure 11] The plan creation device according to any one of Disclosures 1 to 10, wherein the processing plan creation unit creates the processing plan by solving an optimization problem in which a mathematical expression having the evaluation index is an objective function. [Disclosure 12] The process includes transporting the object from a source by a transport means; 12. The plan creation device according to any one of Disclosures 1 to 11, wherein the processing plan is a plan including a timing at which the transportation means departs from the transportation source. [Disclosure 13] The planning device according to Disclosure 12, wherein the processing plan is a plan including timing for placing the same transportation means at each of the multiple transportation sources. [Disclosure 14] 1. A method for generating a treatment plan for treating a plurality of objects, comprising: An acquisition step of acquiring, by an acquisition means, premise information regarding at least attributes of the plurality of objects as premise information for creating the processing plan; an evaluation index calculation step in which an evaluation index calculation means calculates an evaluation index value for evaluating the processing plan using the premise information; a processing plan creation step in which a processing plan creation means creates the processing plan based on an evaluation result of the processing plan using the value of the evaluation index; Equipped with the evaluation index includes an attribute evaluation index for evaluating the degree of similarity of attributes of the plurality of objects; A planning method in which the degree of similarity of the attributes of the multiple objects evaluated using the attribute evaluation index includes the degree of similarity of attributes for all combinations of two objects selectable from the objects to be included in the processing plan. [Disclosure 15] A program for causing a computer to function as each part of the plan creation device according to any one of Disclosures 1 to 13. [Explanation of symbols]

[0100] 110 Processing equipment 111 Acquisition Department 112 Planning Period Division 113 Evaluation Index Calculation Unit 114 Processing Planning Division 115 Output section 120 Input Device 130 Output Device 501~503 Bar graph showing ship loading A~I brand X1, X2 loading port ID i1~i6 Ship ID n First date of the planning period

Claims

1. A planning device for creating a processing plan for processing a plurality of objects, comprising: an acquisition unit that acquires at least premise information regarding attributes of the plurality of objects as premise information for creating the processing plan; an evaluation index calculation unit that calculates an evaluation index value for evaluating the processing plan by using the premise information; a processing plan creation unit that creates the processing plan based on an evaluation result of the processing plan using the value of the evaluation index; Equipped with the evaluation index includes an attribute evaluation index for evaluating the degree of similarity of attributes of the plurality of objects; A planning device in which the degree of similarity of the attributes of the multiple objects evaluated using the attribute evaluation index includes the degree of similarity of attributes for at least some combinations of two objects selectable from the objects to be included in the processing plan.

2. The plan creation device according to claim 1 , wherein the degree of similarity of the attributes of the plurality of objects evaluated by the attribute evaluation index includes the degree of similarity of attributes for all combinations of two objects selectable from the objects to be included in the processing plan.

3. 3. The plan creation device according to claim 1, wherein the evaluation index includes a weighting coefficient for the attribute evaluation index that is based on a difference in processing timing between two objects among the plurality of objects.

4. 4. The plan creation device according to claim 3, wherein the weighting coefficient is a coefficient for weighting a value obtained by squaring the difference.

5. A planning period division unit that divides a planning target period of the processing plan into a plurality of divided periods, The plan creation device according to claim 3 , wherein the difference in the processing timing is expressed as a difference in the divided periods to which the processing timing belongs.

6. 3. The plan creation device according to claim 1, wherein the attribute evaluation indexes include an attribute evaluation index for evaluating a degree of similarity of a type of the object, and an attribute evaluation index for evaluating a degree of similarity of a quantity of the object.

7. The planning apparatus of claim 1 or 2, wherein the processing plan includes plans at one or more locations.

8. The attribute evaluation index is an evaluation index for each location, The plan creation device according to claim 7 , wherein the processing plan creation unit creates the processing plans for the plurality of locations at once by using values ​​of the attribute evaluation indexes for the plurality of locations together.

9. The evaluation index calculation unit groups the objects based on the premise information, 3. The plan creation device according to claim 1, wherein the attribute evaluation index is an evaluation index that evaluates the degree of similarity of attributes of the objects for each of a plurality of groups.

10. 3. The plan creation device according to claim 1, wherein the processing plan creation unit creates the processing plan using a squared value of at least one type of the attribute evaluation index.

11. 3. The plan creation device according to claim 1, wherein the processing plan creation unit creates the processing plan by solving an optimization problem in which a mathematical expression having the evaluation index is used as an objective function.

12. The process includes transporting the object from a source by a transport means; 3. The plan creation device according to claim 1, wherein the processing plan is a plan including a timing for the transportation means to depart from the transportation source.

13. The plan creation device according to claim 12 , wherein the processing plan is a plan including timing for placing the same transportation means at each of the plurality of transportation origins.

14. 1. A method for generating a treatment plan for treating a plurality of objects, comprising: An acquisition step of acquiring, by an acquisition means, premise information regarding at least attributes of the plurality of objects as premise information for creating the processing plan; an evaluation index calculation step in which an evaluation index calculation means calculates an evaluation index value for evaluating the processing plan using the premise information; a processing plan creation step in which a processing plan creation means creates the processing plan based on an evaluation result of the processing plan using the value of the evaluation index; Equipped with the evaluation index includes an attribute evaluation index for evaluating the degree of similarity of attributes of the plurality of objects; A planning method in which the degree of similarity of the attributes of the multiple objects evaluated using the attribute evaluation index includes the degree of similarity of attributes for at least some combinations of two objects selectable from the objects to be included in the processing plan.

15. A program for causing a computer to function as each unit of the plan creation device according to claim 1 or 2.

Images