JPH03243528A - Stowage plan device - Google Patents

Abstract

PURPOSE:To obtain a stowage plan device that stows at a high filling rate a plurality of goods stowing containers by form, by equipping a means conducting grouping and a means operating a mixing arrangement pattern so that a mixing arrangement pattern which has an excellent filling efficiency according to the combination of goods by form may be sought. CONSTITUTION:The plan of stowage is conducted by the control of a deduction control portion 1, and an instruction for an arrangement plan formation is issued from the control portion 1 to an arrangement formation portion 2. Next, at the formation portion 2, combination candidates are sought by applying knowledge within an arrangement knowledge data base portion 5 to goods groupes and stowage surfaces that are not complete with stowage at its stage, and mixing arrangement patterns are calculated with regard to respective candidates. And if there is an excellent arrangement, it is added to an arrangement plan record data base portion 3, and the best arrangement plan among these is adopted, and an advancement for the next arrangement calculation is made. And if there is no excellent arrangement plan, another arrangement plan that is to be kept at the data base portion 3, is adopted, and an advancement for the next arrangement calculation is made, and by repeating this cycle, a stowage plan having an excellent filling efficiency is searched and adopted.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a loading planning device that plans loading when loading cargo using a robot or the like.

(Conventional technology) Conventional loading planning devices load pallets and the like into containers.
Most systems were designed to stow only multiple pieces of cargo of the same size. However, in the recent era of high-mix, low-volume transportation, a system is required for loading multiple cargoes of different shapes, sizes, weights, etc. into a container. In such a system, one stowage requires loading as much cargo as possible into a container in a stable position.

The problem of loading cargo into a container efficiently is basically a combinatorial problem, so for each cargo to be stowed, stowing involves finding all possible loading positions on the container and finding the most efficient loading position among them. In other words, the combination of placement positions with the lowest loading height may be adopted as the solution. However, this method is not practical because the amount of calculation for the number of cargoes and stowage increases as the area of the container increases. Therefore, in reality, instead of finding all combinations, we try some of the combinations by applying rules that are empirically known to yield solutions that are close to the optimal solution, and find the best one among them. A method has been adopted to do so. Rules that are known from experience include, for example, it is better to load tall cargo or cargo with a low surface area first, and it is better to arrange cargo so that the top surface is flat rather than uneven. It is like saying that it is better. In each process of sequentially determining the cargo stowage position, such rules are applied to cargo that has not yet been stowed to determine the best way to stow the cargo.

By repeating this cycle until all cargo is stowed, a stowage plan is created. below,
Japanese Patent Application Laid-open No. 51-206730 using this method using diagrams
A conventional example shown in the publication will be explained.

FIG. 6 is a block diagram of the conventional device, in which (7
) is a placement determining section for stowage, (4) is a product specification data database section, (5) is a placement knowledge base section, and (8) is a placement determining algorithm database section.

In addition, Fig. 7 explains the stowage plan based on the above-mentioned conventional device, and Fig. 7 (a) is a diagram showing the state of cargo arrangement at a certain stage of the stowage plan, in which (21) is a pallet, (
P) and (Q) are cargoes that have already been stowed, that is, stowage is cargo whose position has been determined, (61) is cargo that has been stowed on an empty plane above the container, and (62) and (63) are cargo that has been stowed. In this stowage stage, these three planes are the planes on which the next stowage can be carried out (hereinafter referred to as stowage planes).

In the conventional example, at this stage, the cargo that has not been stowed and the stowage surface are searched for knowledge in the knowledge base that can be applied to the surface condition, and the placement based on the loading method is calculated according to that knowledge, and the most evaluated The cycle of adopting the most expensive item is repeated until the cargo is exhausted.

Furthermore, Za diagram shows the length, width, height, and stowage of cargo P, Q, R, and S, and the number of stowage pieces, and Fig. 9 shows the flow of processing by the arrangement determining section (A) for stowage in the conventional example. FIG.

Hereinafter, taking the cargo stowage shown in FIG. 8 as an example, the operation of the stowage arrangement determination section (7) will be explained in the order of the steps in the flowchart shown in FIG. 9.

Step 591 (extraction of characteristics of cargo arrangement): The stowage arrangement determination unit (7) calculates the characteristics of cargo arrangement at a certain stowage stage. Cargo placement characteristics include the width, length, height, and cargo volume of cargo that has not yet been stowed (hereinafter referred to as unloaded cargo); This is a value such as height. In the arrangement situation shown in Figure 7(a), for unloaded cargo R and S, the loading is on the surface (61),
The feature amounts for (62) and (63) are calculated.

Step 592 (Creation of Alternative Plans) Judging from the characteristics calculated in Step 591, several combinations of candidates for the next cargo to be stowed and candidates for the stowage surfaces to be stowed are created. In the layout knowledge base section (5), knowledge (Fig. 8) that associates the values of various characteristics with the stowage method is described, and based on the characteristics calculated in step 591, the above combination plan is created according to this. (hereinafter referred to as an alternative). 7th
In the case of Figure (a), two alternatives are created: loading the cargo R in a rectangular shape on the surface (61), and loading the cargo S in a rectangular shape on the surface (61).

Step 593 (Location Calculation) For each alternative selected in step S92, the stowage calculates the individual cargo placement in the three-dimensional space on the container. In determining the arrangement at this time, a plurality of arrangements that are considered to have good filling efficiency are calculated according to the algorithm written in the arrangement determination algorithm database section (8). Figure 7 (b), (c), (d)
Three configurations have been calculated.

-Step 594 (evaluation of placement results) 2nd step S93
Select the most desirable arrangement from among the arrangements calculated in , and update the cargo arrangement status.

Step 595 (completion determination): If the arrangement of all cargoes is determined, the process ends; otherwise, the process returns to step 591.

[Problem to be solved by the invention]

Since conventional load planning devices are configured as described above, there is proof that even if a load placement plan judged to be the best is selected in a certain process, the optimal load placement result will ultimately be obtained. The problem is that there is no. Furthermore, when creating one layout for one type of cargo and one stowage, the combination of surfaces is considered, and the arrangement of multiple types of cargo at the same time is not considered. Considering placement patterns that combine cargo at the same time has the potential to obtain high filling efficiency, but as a matter of course, the number of combinations increases dramatically.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a loading planning device that can load a plurality of different types of cargo horizontally onto a container with high filling efficiency.

(Means for Solving the Problems) A loading planning device according to the present invention includes a means for grouping a plurality of types of cargo, a means for calculating a mixed arrangement pattern based on a combination of the grouped plurality of types of cargo, and a mixed arrangement. A placement room generation unit that has a means to generate multiple placement rooms based on a pattern, a placement plan history data unit that has past placement rooms as data, and a product specification that has container specifications as data for cargo and stowage. The system is equipped with a data storage unit, a placement knowledge base unit that has know-how regarding cargo placement as data, an evaluation algorithm data storage unit that has data storage related to evaluation algorithms, and an inference control unit that generates a stowage plan.

(Operation) In this invention, according to the above-mentioned configuration, the stowage is performed by grouping a plurality of cargoes of different shapes in the process of plan generation, and creating a mixture obtained by combining the plurality of cargoes included in the group. In addition to calculating the placement pattern,
Stowage is a process of plan generation, and when a decline in filling efficiency occurs, part of the stowage plan is discarded and a plan is made for another placement room.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. For convenience of explanation, the stowage will be explained using pallets as containers.

In FIG. 1, (1) is an inference control unit for generating a stowage plan, (2) is an arrangement room generation unit, (3) is an arrangement plan history database unit, (4) is a product specification database unit,
(5) is the knowledge database section for stowage, and (6) is the evaluation algorithm database section.

Figure 2 shows the arrangement of pallets and cargo at a certain stage of stowage planning; figure (a) shows the arrangement at the current stage, and figures (b), (c), and (d) is the same figure (
This figure shows the three cargo compartments that should be placed next in the situation a). Furthermore, FIG. 3 shows the data structure of the layout plan history database section (3).

The operation related to the above configuration will be explained. First, we will give an overview of stowage plan generation. The stowage planning is carried out under the control of the inference control section (1). First, an instruction to generate a placement room is issued from the inference control section (1) to the placement room generation section (2). In the placement room generation unit (2), the knowledge in the placement knowledge database unit (5) is applied to the groups and stowage planes of cargoes that have not yet been stowed at that stage, and combination candidates are obtained, and each A mixed placement pattern is calculated for the candidates. If there is a placement with a high evaluation, it is added to the placement plan history database section (3), the best placement room is selected from the added placement rooms, and the process proceeds to the next placement calculation. If there is no highly rated placement room, the placement plan history database department (3
) and proceed to the next placement calculation. By repeating this cycle, we search for a stowage plan with high filling efficiency.

Next, the layout plan history database section (3) shown in Figure 3
Explain the structure of This history database has a tree structure. A wooden salt node indicates one placement room. In one placement room, the type of cargo to be placed, the type of placement pattern, placement position, placement direction, placement efficiency, etc. are described. Arrangement generation start pointer (31) (hereinafter referred to as pointer (3)
1) always points to the latest placement room. That is, as a result of calculating the placement rooms in the placement room generation unit (2), if there are placement rooms with high evaluations, these are added in parallel as child nodes of the placement room pointed to by the pointer (31) in the order of the highest evaluation. Then, the pointer (31) points to the placement room added at the beginning (that is, the placement room with the highest evaluation). If there is no placement room with a high evaluation, the placement plan pointed to by the pointer (31) is discarded, and the placement plan with the next highest rating after this placement plan, which was added in parallel with the discarded placement plan, is moved with the pointer (31). Instruct.

Furthermore, the structure of data called cargo placement status will be explained. One stowage plan consists of a plurality of layout plans, and can be obtained from the layout plan history database section (3). Tracing the tree vertically from the placement plan pointed to by the pointer (31) (following the parent notes in sequence), and integrating the cargo placement information included in the placement plans followed until reaching the root note, Information on the location of all cargo is required. This placement information is called cargo placement status.

FIG. 4 shows the length, width, height, and number of cargoes A-E, and FIG. 5 is a flowchart showing the processing flow of this apparatus. Hereinafter, the operation will be explained in the order of steps in the flowchart shown in FIG. 5, taking the cargo stowage shown in FIG. 4 as an example.

Step 551 (creation of cargo placement data): Upon receiving an instruction to generate a placement plan from the inference control section (1), the placement plan generation section (2) first creates data on the cargo placement situation at that time. Next, from this data, the characteristics of cargo arrangement, that is, the number, width, length, height, and volume of unloaded cargo, and for loading, the width, length, height, and height of adjacent cargo. Extract differences etc. In the case of Figure 2 (a), cargo A is loaded and four types of unloaded cargo B
.

Since surfaces (21) and (22) (indicated by diagonal lines) exist in CD and E, the feature amounts related to these surfaces are calculated.

Step 552 (Grouping of unloaded cargo) 2. Similar cargo groups are created for unloaded cargo that have the same length, width, and height at least one of the same. A similar cargo group is a combination of cargoes that may form a mixed arrangement pattern with high filling efficiency. Of course, one type of cargo may belong to multiple groups. Figure 2(a)
In this case, cargoes B and C are in one group, and cargoes C and D are in one group.
form two groups.

Step 553 (Extraction of combination candidates): Step 55
The placement knowledge database part (5
), and the cargo and stowage are determined by the combination of surfaces. In the arrangement knowledge database section (5), rules are described that have feature quantities in the condition section and stowage methods in the execution section.

In the case of the figure, one of the matching rules is as follows.

if (Cargo, loading end flag - not loaded) & (Cargo b height = high) & (Cargo, total volume > normal) & (Loading is on surface p, height = low) & (Loading is on surface p, width = wide) & then make (Combination candidate A, cargo.

When the rule is executed, cargo that satisfies the conditions of this rule, data containing three types of values: area for loading and method for loading (combination candidates) are executed. ) is created. Figure 2 (a
), cargo B, loading is by plane (22), and loading is by method W.
1 and are extracted as one combination candidate.

Step 554 (creation of placement plan candidates) 2nd step S53
For each of the combination candidates extracted in step 1, the stowage determines the three-dimensional placement position on the container. The similar cargo group created in step S52 is examined, and groups in which the cargo described in the combination candidate exists are extracted among the cargoes included in the group. Find multiple mixed placement patterns that combine cargo.

'For the mixed arrangement pattern, first of all, a plurality of efficient three-dimensional three-dimensional arrangement patterns for a single type of cargo are determined. In this case, being efficient means that there are few gaps between pieces of cargo, that the same number of pieces of cargo are arranged up to the top tier, or that there is no excess cargo on the top tier. Second, all possible combinations of placement patterns, that is, mixed placement patterns, are determined by combining the respective placement patterns. Each of the plurality of mixed placement patterns created in this way becomes a final placement plan candidate (hereinafter referred to as a placement plan candidate). That is, the combination candidate plus the information on the mixing device pattern becomes the placement plan candidate.

Step 555 (determination of placement plan): Calculate placement efficiency for a plurality of placement plan candidates, and select the plurality of placement plan candidates with high evaluation values as placement plans. The placement efficiency is calculated by adding empirical weights to, for example, the amount of wasted space on the pallet created by arranging the cargo, and the area of the newly created loading surface. In this example, the arrangement of cargo B or C shown in FIGS. 3(b), (c), and (d) is adopted as the arrangement plan. The plurality of placement plans generated in this way are delivered to the inference control unit (1).

Step 556 (placement plan history database update) The inference control unit (1) adds the plurality of placement plans selected in step S55 to the placement history data structure diagram. That is, in the placement plan tree, branches are increased by the number of new placement plans to be added beyond the placement plan pointed to by the placement plan generation start pointer (31), and each placement plan is added in order of highest evaluation value. However, before adding, check whether the placement business you are adding is the same as the placement plan you created previously.
Add only if they are not the same.

Step 557 (Change placement plan generation start pointer) - If the best evaluation value among the evaluation values of the placement plans added in step 556 is higher than the determined threshold, the placement plan with this evaluation value is adopted and indicated with a pointer (31). When the best evaluation value is smaller than the threshold, the plan is backtracked. That is, the placement plan with the next best evaluation value is selected from among the placement plans having the same parent as the placement plan pointed to by the pointer (31), and pointed to by the pointer (31). If the next placement plan does not exist, the tree of placement plans is further traced and the same process is performed.

Step 558 (completion judgment): If there is no unloaded cargo remaining, the plan generation process is ended; otherwise, the process returns to step 551. After completing the plan generation process, the placement history database unit (5) obtains data on the cargo placement status from the placement plan pointed to by the pointer (31).
Stowage requires planning.

In the above embodiment, the case where a plan is obtained for one stowage is explained, but it goes without saying that it is also possible to obtain a plan for a plurality of stowages.

〔Effect of the invention〕

As described above, according to the present invention, there is provided a layout plan history database section that has past loading layout plans as data, and a system for determining a mixed layout pattern with good filling efficiency formed by combining different types of cargo. Since it is equipped with means for grouping and means for calculating a mixed arrangement pattern, it is possible to carry out a loading plan for loading multiple cargoes of different shapes onto containers with high filling efficiency with a relatively small amount of calculation. It has the effect of being able to make plans.

[Brief explanation of drawings]

Figure 1 is a block configuration diagram of a loading planning device in an embodiment of the present invention, Figures 2 (a) to (d) are explanatory diagrams of the layout status at each stage of loading planning, and Figure 3 is a history of layout plans. Data structure diagram of database section (3), Figure 4 shows cargo A-H
FIG. 5 is a flowchart showing the flow of processing in this embodiment, FIG. 6 is a block diagram of the conventional device, and FIG.
Figures (a) to (d) are explanatory diagrams of the layout of cargo at each stage of planning to explain the conventional equipment, and Figure 8 is an explanatory diagram of the cargo P~
FIG. 9, which is an explanatory diagram of S, is a flowchart showing the flow of processing in a conventional example. In addition, the same reference numerals in the figures refer to the same or corresponding parts. Ward Marunouchi 2-2-3 Name
Name (601) Moriya Shiki, Representative of Mitsubishi Electric Corporation

Claims (1)

[Claims] A placement plan that includes means for grouping multiple types of cargo, means for calculating a mixed placement pattern by combining the grouped multiple types of cargo, and means for generating multiple placement plans based on the mixed placement pattern. a generation section, a layout plan history database section that has past layout plans as data, a product specification database section that has cargo and stowage container specifications as data, and a layout knowledge base section that has know-how regarding cargo placement as data; an evaluation algorithm database department having a database regarding evaluation algorithms;
A loading planning device comprising an inference control unit that generates a loading plan.