JP7735973B2 - Analysis processing device and analysis processing method - Google Patents

Description

The present invention relates to technology for analyzing the ability of workers to perform waste sorting tasks.

Patent Document 1 discloses a waste sorting method in which brought-in waste is separated into oversized and undersized waste using a large sieve, and each is transported by a large oversized conveyor and an undersized conveyor, respectively. Large items are then manually sorted from the oversized waste by type by workers waiting on the conveyor's transport path, and the items are then dumped into a discharge chute that opens next to the workers.

Japanese Patent Application Publication No. 11-19593

When workers sort waste by item, it is desirable to be able to understand their capabilities, as this allows workers to be appropriately assigned to one of several sorting tasks.

The object of the present invention is to provide technology that can grasp the ability of workers who sort waste by item.

In order to solve the above problem, one aspect of the present invention provides an analysis processing device that analyzes the work ability of a worker performing sorting work in a sorting work in which pre-sorting waste, which is a mixture of multiple items, is converted into post-sorting waste, which is sorted by item, and includes: an acquisition unit that acquires pre-sorting waste information including the weight of the pre-sorting waste before sorting and the items contained in the pre-sorting waste; acquires sorting work information including information on the worker engaged in the sorting work of the pre-sorting waste and the sorting work time performed to perform the sorting work ; and acquires post-sorting waste information indicating the amount of waste by item after sorting the pre-sorting waste; and an analysis unit that derives sorting work parameters indicating the work ability of the worker for each worker based on the pre-sorting waste information, sorting work information, and post-sorting waste information. The analysis unit derives the processing speed for each waste item as the sorting work parameter for the worker.

Another aspect of the present invention is an analytical processing method. This method analyzes the work capacity of a worker performing a sorting task in which pre-sorting waste, which is a mixture of multiple items, is sorted into post-sorting waste, which is sorted by item, and each step is executed by a computer. The method includes the steps of: acquiring pre-sorting waste information including the weight of the pre-sorting waste before sorting and the items contained in the pre-sorting waste; acquiring sorting work information including information about the worker engaged in the sorting task and the sorting work time spent performing the sorting task; acquiring post-sorting waste information indicating the amount of waste by item after sorting the pre-sorting waste; and deriving sorting work parameters indicating the work capacity of the worker for each worker based on the pre-sorting waste information, sorting work information, and post-sorting waste information. In the deriving step, the sorting work parameter for the worker is a processing speed for each waste item.

This invention provides technology that can assess the capabilities of workers who sort waste by item.

FIG. 1 is a diagram for explaining an overview of waste disposal. FIG. 2 is a diagram illustrating a functional configuration of the analysis processing system. FIG. 10 is a diagram illustrating an example of sorting task parameters of a worker. 10 is a flowchart of a process for deriving sorting task parameters for each worker.

Figure 1 is a diagram illustrating an overview of waste disposal. Each vehicle 10 delivers pre-sorted waste 12a, pre-sorted waste 12b, and pre-sorted waste 12c (referred to as "pre-sorted waste 12" when not distinguishing between them) to a disposal facility 14. Pre-sorted waste 12 is a mixture of multiple items, and the items may vary depending on the supplier. For example, pre-sorted waste 12a includes glass scraps, wood chips, and rubble, pre-sorted waste 12b includes rubble and scrap metal, and pre-sorted waste 12c includes waste paper and plastic. Suppliers declare to facility staff at the disposal facility 14 what the waste contains.

At the processing facility 14, multiple workers receive the pre-sorted waste 12 and manually sort the pre-sorted waste 12 that has been conveyed down the conveyor by item. The pre-sorted waste 12 is transformed into post-sorted waste through the sorting process and delivered to each facility. For example, rubble is delivered to the landfill 16, wood waste and paper waste are delivered to the incineration plant 18, and glass scraps, metal scraps, and plastics are delivered to the recycling facility 20. In this way, pre-sorted waste containing a mixture of multiple wastes is brought to the processing facility 14, sorted by item, and delivered to the appropriate disposal site. Plastics can also be delivered to the incineration plant 18 as well as the recycling facility 20.

The abilities of workers engaged in sorting work vary. For example, some workers are not good at sorting large items such as rubble but are good at sorting cardboard and paper scraps, while others are fast but have low sorting accuracy. It is preferable to assign workers to appropriate sorting tasks based on their abilities. Therefore, the analysis processing device of this embodiment collects waste information and sorting work information before and after sorting, and derives the sorting work ability of each worker, i.e., the sorting work parameters for each worker.

Figure 2 is a diagram showing the functional configuration of analysis processing system 1. Each function of analysis processing system 1 can be configured in hardware using circuit blocks, memory, and other LSIs, and in software, it is realized by system software or application programs loaded into memory. Therefore, those skilled in the art will understand that each function of analysis processing system 1 can be realized in various ways using hardware alone, software alone, or a combination of these, and is not limited to any one of these.

The analysis processing system 1 comprises an input unit 22, an external terminal device 24, an analysis processing device 26, and an output unit 28. The input unit 22 may be, for example, a touch panel, keyboard, microphone, etc., and receives pre-sorting waste information, post-sorting waste information, or sorting work information, and sends this information to the analysis processing device 26. The external terminal device 24 is also connected to the analysis processing device 26 via a network and sends pre-sorting waste information, post-sorting waste information, or sorting work information. In other words, pre-sorting waste information, post-sorting waste information, and sorting work information are sent from the input unit 22 or the external terminal device 24.

The pre-sorted waste information includes the delivery destination ID, the pre-sorted waste ID, the weight of the pre-sorted waste, the waste items contained in the pre-sorted waste, and the proportion of each item in the pre-sorted waste. The waste items contained in the pre-sorted waste and the proportion of each item in the pre-sorted waste may be determined by the supplier's declaration. For example, the supplier declares that the pre-sorted waste 12a shown in Figure 1 is composed of 30% glass waste, 20% wood chips, and 50% rubble.

The sorted waste information includes the delivery destination ID, the type of sorted waste, and the weight of each type of waste. The sorted waste information is obtained by measuring the weight at the processing facility 14 after sorting and inputting it by the worker.

Sorting work information includes the sorting work ID, the ID of the worker engaged in the sorting work, the ID of the unsorted waste to be worked on, the sorting work time, the work area of the area where the sorting work is performed, sorting accuracy, recycling rate, and waste layer thickness. Information related to the execution of sorting work includes the sorting work time, the work area of the area where the sorting work is performed, sorting accuracy, recycling rate, and waste layer thickness. Sorting accuracy is calculated when an inspector checks the sorted waste and removes impurities, and may be calculated from the weight of the impurities and the sorted waste. The waste layer thickness is determined by the height of the gates installed on the belt conveyor. Because unsorted waste passes through gates installed on the belt conveyor, the waste layer thickness can also be said to be determined by the thickness of the unsorted waste. The sorting work time may be measured using camera images monitoring the sorting work. Sorting work information may be obtained through input by workers or inspectors.

The output unit 28 is a display, speaker, or the like, and outputs the information generated by the analysis processing device 26. The analysis processing device 26 includes an acquisition unit 30, an analysis unit 32, a calculation unit 34, a reception unit 36, a memory unit 38, and a determination unit 40.

The acquisition unit 30 acquires pre-sorting waste 12 information, which includes the weight of the pre-sorting waste 12 before sorting and the items contained in the pre-sorting waste 12. The acquisition unit 30 acquires sorting work information, which includes information about the workers who were engaged in sorting the pre-sorting waste 12 and information about the execution of the sorting work. The acquisition unit 30 acquires post-sorting waste information, which indicates the amount of waste for each item after sorting the pre-sorting waste 12. In other words, the acquisition unit 30 acquires pre-sorting waste information, sorting work information, and post-sorting waste information.

The analysis unit 32 derives sorting work parameters for each worker based on the pre-sorting waste information, sorting work information, and post-sorting waste information. The analysis unit 32 may derive the sorting work parameters using a multiple regression analysis method, or may derive the sorting work parameters using a machine learning method. The sorting work parameters indicate the worker's skill when performing the sorting work, and are derived from past work results.

However, because sorting work is performed by a team of multiple workers, it is difficult to measure the ability of each individual worker. The analysis unit 32 collects multiple results of sorting work performed by the team of workers and derives the ability of each worker based on the differences between them. For example, the analysis unit 32 compares the sorting work performed by "Worker A" and "Worker B" with the sorting work performed by "Worker A," "Worker B," and "Worker C" to measure the ability of "Worker C." In this way, the analysis unit 32 derives the sorting work parameters of each worker based on the differences between the workers engaged in the sorting work. However, because the content of the sorting work varies, the analysis unit 32 cannot derive the sorting work parameters based on the differences alone, and must instead use techniques such as multiple regression analysis.

The sorting parameters for each worker include the processing speed (kg/hour) for each item and the recycling rate for each item. The recycling rate refers to the sorting accuracy for each item. Sorting accuracy is determined by an inspector's survey, who determines the proportion of mixed items in the sorted waste. Waste containing only plastic is delivered to the recycling facility 20, while mixed waste containing wood chips and plastic is delivered to the incineration plant 18. Therefore, the higher the sorting accuracy, the higher the recycling rate. Simply put, the sorting processing speed is calculated from the weight of the waste before sorting, the working time, and the number of workers. However, because workers and items are mixed, the processing speed for each item of a specific worker is calculated using multivariate analysis.

The sorting work parameters for each worker may further include overall sorting accuracy, processing speed depending on the work area, etc. As the work area becomes larger, the amount of movement during work increases, which can result in significant changes in processing speed depending on the worker.

The analysis unit 32 may use a program that inputs pre-sorting waste information, sorting work information, and post-sorting waste information, and outputs the processing speed for each worker by item. This program may be created using multiple regression analysis, or may be created using machine learning or deep learning techniques.

Figure 3 shows an example of sorting work parameters for workers. The vertical axis in Figure 3 is skill value, which is a coefficient of processing speed, and multiplying the skill value by a predetermined processing speed gives the worker's processing speed. Figure 3 shows the skill values of "Worker A," "Worker B," and "Worker C."

For example, Worker A's cardboard skill value of 42 is higher than Worker B's cardboard skill value of 44, but lower than Worker C's cardboard skill value of 46. Furthermore, Worker A's metal skill value of 48 is higher than Worker B's metal skill value of 50, and higher than Worker C's metal skill value of 52. This shows that Worker A has few weak points, while Worker C has a very wide range of strong and weak points.

In this way, the sorting work parameters of each worker specifically indicate which worker is suitable for which sorting work. The analysis unit 32 may use machine learning techniques to learn and update the sorting work parameters of each worker based on the pre-sorting waste information and post-sorting waste information. The analysis unit 32 associates the sorting work parameters of each worker with the worker ID and stores them in the memory unit 38.

Returning to Figure 2, the calculation unit 34 creates multiple patterns of worker teams made up of multiple workers. In other words, the calculation unit 34 creates multiple combinations of workers that make up a worker team. The number of people in a worker team is two or more.

The calculation unit 34 calculates the correlation between the worker team and the objective of the sorting work based on the pre-sorting waste information and the sorting work parameters for each worker. The objective of the sorting work is, for example, processing speed, work completion time, recycling rate, sorting accuracy, weight of waste after sorting, etc., and is an indicator of the results of the sorting work. The objective of the sorting work may be the pre-sorting waste information for the requested sorting work. The correlation is a predicted value of the work results when the sorting work is performed. In other words, the calculation unit 34 calculates, for each worker team, a predicted performance value of the sorting work according to the objective of the sorting work based on the pre-sorting waste information and the sorting work parameters for each worker. The predicted performance values include processing speed, work completion time, recycling rate, and sorting accuracy. In this way, by having the calculation unit 34 calculate in advance the predicted performance values of various worker teams composed of combinations of workers, i.e., the capacity values for the sorting work, personnel allocation and work plans according to the sorting work can be easily created.

The calculation unit 34 may calculate sorting work parameters for each worker team based on the pre-sorting waste information and the sorting work parameters for each worker. The sorting work parameters for the worker team calculated here may be the processing speed for each item and the recycling rate for each item. The sorting work parameters for each worker team calculated by the calculation unit 34 are stored in the memory unit 38. The memory unit 38 stores the worker IDs included in the worker team in association with the sorting work parameters.

The determination unit 40 determines the worker teams to be engaged in the sorting work based on the predicted execution values of the sorting work for each worker team. This allows the determination unit 40 to appropriately select the worker teams to perform the sorting work. The determination unit 40 determines the worker teams according to objectives such as improving processing speed, work completion time, recycling rate, and sorting accuracy. For example, the determination unit 40 may determine the worker teams with the objectives of high processing speed and limited number of workers, or may determine the worker teams with the objectives of high recycling rate and limited number of workers. The determination unit 40 may also determine the worker teams with the objectives of increasing the recycling rate while efficiently improving processing speed and number of workers.

The analysis unit 32 performs a process to associate pre-sorted waste information with post-sorted waste information, and the memory unit 38 stores the pre-sorted waste information in association with the post-sorted waste information. Post-sorted waste is often a mixture of multiple pre-sorted wastes, and post-sorted waste and pre-sorted waste cannot be directly associated. Pre-sorted waste information includes the weight of the pre-sorted waste and the proportion of each item contained in the pre-sorted waste. Therefore, the weight of each item of pre-sorted waste is simply calculated by multiplying the weight of the pre-sorted waste by the proportion of the item. The weight of each item of pre-sorted waste calculated in this way does not match the weight of each item of post-sorted waste. Therefore, the analysis unit 32 assumes the weight of each item of post-sorted waste to be the correct value, and performs a process to back-calculate the weight of each item of post-sorted waste and apply it to multiple pieces of pre-sorted waste information to estimate the weight of each item of pre-sorted waste. This allows suppliers who deliver pre-sorted waste to understand the actual sorting results.

The memory unit 38 stores the sorted waste information in association with delivery destination information for the sorted waste. Because the sorted waste information is associated with the pre-sorted waste information, the supplier that delivered the pre-sorted waste can determine where the sorted waste has been delivered. The information stored in the memory unit 38 can be output from the output unit 28 in response to instructions from the input unit 22 and external terminal device 24.

Figure 4 is a flowchart of the process for deriving sorting work parameters for each worker. The acquisition unit 30 acquires pre-sorted waste information from the input unit 22 or external terminal device 24 (S10). Workers work in teams to perform sorting work on pre-sorted waste, and a camera and/or inspector monitors the sorting work (S12) and generates sorting work information. The inspector inputs the sorting work information, and the acquisition unit 30 acquires the sorting work information from the input unit 22 or external terminal device 24 (S14).

The inspector measures the weight of the sorted waste for each item, measures the degree of sorting for each item, and generates sorted waste information (S16). The acquisition unit 30 acquires the sorted information from the input unit 22 or the external terminal device 24 (S18).

The analysis unit 32 executes an analysis process that inputs pre-sorting waste information, sorting work information, and post-sorting waste information and outputs sorting work parameters for each worker (S20). The memory unit 38 stores the sorting work parameters for each worker in association with the worker ID (S22).

Please note that the examples are merely illustrative, and those skilled in the art will understand that various variations in the combination of components are possible and that such variations are also within the scope of the present invention.

1 Analysis processing system, 10 Vehicle, 12 Pre-sorted waste, 14 Processing facility, 16 Landfill disposal site, 18 Combustion plant, 20 Recycling facility, 22 Input unit, 24 External terminal device, 26 Analysis processing device, 28 Output unit, 30 Acquisition unit, 32 Analysis unit, 34 Calculation unit, 36 Reception unit, 38 Memory unit, 40 Determination unit.

Claims (3)

An analytical processing device that analyzes the work ability of a worker who performs a sorting operation in a sorting operation in which pre-sorted waste, which is a mixture of multiple items, is sorted into sorted waste, which is sorted for each item,
an acquisition unit that acquires pre-sorting waste information including the weight of the pre-sorting waste before sorting and the items contained in the pre-sorting waste, acquires sorting work information including information on workers who engaged in sorting the pre-sorting waste and the sorting work time spent performing the sorting work , and acquires post-sorting waste information that indicates the amount of waste for each item after sorting the pre-sorting waste;
an analysis unit that derives sorting work parameters that indicate the work ability of each worker based on the pre- sorting waste information, the sorting work information, and the post- sorting waste information,
The analysis processing device is characterized in that the analysis unit derives a processing speed for each waste item as the sorting work parameter of the worker . 2. The analysis processing device according to claim 1 , further comprising a storage unit that stores the pre -sorted waste information and the sorted waste information in association with each other , and that stores the sorted waste information in association with delivery destination information for the sorted waste. 1. A method for analyzing the work ability of a worker who performs a sorting operation to convert pre-sorted waste, which is a mixture of multiple items, into sorted waste, which is sorted by item, and performing each step of the analysis processing by a computer, comprising:
A step of acquiring pre-sorted waste information including a weight of the pre-sorted waste before sorting and items contained in the pre-sorted waste;
A step of acquiring sorting work information including information on workers who have engaged in sorting work of unsorted waste and the time spent on sorting work ;
A step of acquiring post-sorting waste information indicating the amount of waste for each item after sorting the pre-sorting waste;
and deriving a sorting work parameter indicating the work ability of each worker based on the pre- sorting waste information, the sorting work information, and the post- sorting waste information ,
The analytical processing method is characterized in that in the deriving step, a processing speed for each waste item is derived as the sorting work parameter of the worker .