JP2003022119A - Multiproduct production system, its design/operation method, its design/operation program and recording medium recorded with the program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To estimate the number of needed personnel accompanying a production class changeover work and an initial setup work in order to reduce production cost. SOLUTION: A first class product is processed by a certain shared device. When a buffer becomes empty, a processing of the first class product is interrupted and a second class product is processed. When the second type buffer becomes empty, a third class product is processed. While the third class product is processed, the first class product is continuously stored in the buffer and the second class product is newly stored in the buffer as well. Namely, a plurality of other class products are proceeds during a period to store the first class product. The device is constituted so that the processing of the first class product is restarted when quantity to be stored in the buffer reaches allowable quantity Nmax for processing the product. When other class product reaches Nmax of the product before the first class product reaches Nmax, the product may be processed as the first class product.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multi-product production system, a design / operation method thereof, a design / operation processing program thereof, and a recording medium recording the program. In particular, the present invention provides a method for designing and operating a production system and specifically discloses the system, and belongs to a technical field called system engineering or management engineering. In the embodiments of the present invention, a semiconductor integrated circuit production system will be described in detail as an example, but the present invention is not limited to this and can be applied to various systems.

In the present invention, in order to reduce the production cost when producing a large number of products in a single production line as much as possible, the amount of mechanical equipment to be installed in the production system,
It is applied when designing the number of personnel required to operate the production equipment and optimizing the operation method.
In particular, when operating a large number of product types, the technical field that calculates the number of personnel involved in the work of switching the device according to the specifications and production amount of each product, so-called setup work, and the setup work are specifically operated. Belongs to the technical field.

[0003]

2. Description of the Related Art Up to now, several proposals have been made as a design system for this kind of equipment quantity and number of personnel. Related Materials (1) (Kazuyuki Saito, 'Optimization of th
e Number of Machines and Operators Required for LS
I Production ', IEICE TRANS.ELECTRON., VOL. E79-C,
No.8, pp1112-1119, August 1996) provides a method for designing the number of machines and equipment and the number of personnel involved in semi-automatic processing. In the semi-automatic process, the processing time in each process is
It is classified into five time elements: pre-processing standby time, pre-processing time, automatic processing time, post-processing standby time, and post-processing time.
It is premised that the product is manually transported from the input buffer to the device in the pre-treatment, and the product is manually transported from the device to the output buffer or the input buffer in the next process in the post-treatment. The product delivered to the device is automatically processed for a predetermined time. According to the related material (1), it was possible to predict the value close to the actual value by calculating the equipment and the number of personnel of the production line of almost a single product that produces a small number of products.

However, related materials (2) (K. Saito, S. A.
rima, T. Yokoyama, and H. Yamanaka, 'Application
of a Resource Planning System for a VLSI Assembly
Facility ', ISSM'99, pp 345-347, 1999, Santa Cl
(See ara), it is stated that in the case of a factory that handles a wide variety of products, the effective operating rate of the device is lower than the value predicted based on the related document (1). For this reason, the related document (2) describes a method of determining the amount of equipment in consideration of the effective decrease in the operating rate and anticipating the effective decrease. A tuning value was introduced as a parameter that represents a decrease in the operating rate, and the accuracy of calculating the amount of equipment in each process of production,
There was an improvement in the prediction of production time (TAT, Turnaround Time).

[0005]

Such a tuning value corresponds to a decrease in the facility operating rate due to the production of various products, but is an empirical value in the production line. In other words, there is no method provided to determine the tuning value when the number of products handled on the line, the production amount of each product, or the specifications are significantly changed, that is, the processing time is changed due to the difference in specifications. . Further, what can be provided by this method is only a prediction regarding the required amount of equipment at the time of multi-product production, and in particular, there remains a problem that the accuracy of predicting the number of personnel involved in multi-product production remains low.

Each product type produced on a single line has its own set delivery date, and therefore each process has its own stayable time. The stayable time in each process is the sum of the “waiting time” from the time when the product reaches the process to the start of the process and the “service time” from the start to the end of the process. Service time is an essential time to meet product specifications. Therefore, the process stay time is adjusted by adjusting the waiting time. In a line that handles a large number of products, it is necessary to increase the operating rate of the device in order to keep the delivery time and production volume of each product while adjusting the waiting time and at the same time keep the production cost as low as possible.

Previously provided method (related material (3): S. Nakamura et al., 'Simulation System fo
r Resource Planning and Line Performance Evaluatio
n of ASICManufacturing Lines', IEICE TRANS. ELECT
RON., VOL. E79-C, NO.3, pp290-300, March 1996.)
In consideration of compliance with the delivery date of each product, there is no consideration regarding the operating rate of the equipment. Generally speaking, when trying to comply with the delivery date of each product, the number of product switching is increased, which means that the operation rate of the device is reduced.

Therefore, in a multi-product production system, it is necessary to establish a line operation method that satisfies both contradictory requirements of delivery date compliance and improvement of operating rate. Finally, it is necessary to clarify the operation standard for the setup work that accompanies the product type change after the delivery time of all the product types is satisfied.

In view of the above points, the present invention aims to reduce the production cost in a multi-product type production line, and it is necessary to perform product type switching and setup work in the line, which has been difficult to handle by the conventional technology. The purpose is to provide a method for estimating the number of personnel. A further object of the present invention is to provide a method for estimating the required mechanical equipment amount in consideration of the amount of decrease in the equipment operating rate due to the setup work. Further, the present invention, by these methods,
It is an object of the present invention to disclose a method of designing a production system for performing multi-product production with minimum equipment and personnel. Further, the present invention discloses an operation method of a multi-product production system that achieves both delivery time compliance and improvement of operation rate of each product by utilizing the equipment and personnel designed in this way, and the operation method can be realized. The purpose of the present invention is to disclose the production system.

[0010]

[Means for Solving the Problems] As a first embodiment, a management method for observing the delivery date of a product is disclosed.
The method shown here installs an input buffer for each product type, and sets the allowable amount in that buffer to keep the amount of the product below the allowable value. It keeps the time below the allowable value. The management of the allowable amount can be easily realized by using the existing sensor, which is effective in automating the production system.

As a second embodiment, a periodical allocation method is disclosed as an allocation method for each kind when the setup work is taken into consideration. With this allocation method, it is possible to easily find a solution when allocating two types of products to one device, and from that result, the amount of equipment and the number of personnel that can operate the system without shortage due to the production of multiple types. It is possible to design the required upper limit value of. As a result, it is possible to suppress excessive capital investment and personnel recruitment, and reduce production costs.

As a third embodiment, a concrete operation method in the case where an input buffer is provided for each product type has been shown. In particular, the method of determining the priority of the product type to be processed is disclosed from the in-process amount of the product in the buffer. By this operation method, it became clear that the apparatus can secure an operation rate close to 100% if the setup time is ignored. In addition, by properly managing the amount of work in progress, it became possible to create equipment downtime, and it became possible to schedule maintenance inspections that period. That is, it is possible to reduce the production cost by improving the operation rate of the device, and it is possible to improve the reliability of the production system by scheduling the maintenance and inspection.

Further, the embodiment discloses the configuration of the production system. Although a part of human labor is required for the setup work, it is possible to configure a semi-automatic system, reduce the number of process monitoring personnel, and reduce the production cost.

According to the first solution of the present invention, a buffer for inputting each product type is provided in a product processing process in a multi-product manufacturing system, and the time when the product type stays in the buffer can stay in the process. Provide a method for designing and operating a multi-product production system, characterized in that a control value of a buffer is set so as to be within an average allowable time.

According to the second solving means of the present invention, it is constituted by a plurality of devices, and a maximum of two kinds are assigned to one device in a processing process for processing a plurality of products.
Calculate the allocation cycle for the device to which the product type is allocated,
A design / operation method of a multi-product production system characterized by designing the number of personnel required for setup work based on the cycle.

According to the third solution of the present invention, the number of products in the buffer is periodically monitored in the processing step composed of a plurality of devices and an input buffer provided for each product type. An arrival rate estimated from past results, a service rate of processing of the product type, and a management value of the buffer are used to determine the priority of the product type to be processed and allocate it to the device. design·
Provide an operation method.

According to the fourth solving means of the present invention, the process management computer inputs the preset input data, and the process management computer executes the corresponding process based on the production plan of each product type. a step of calculating the arrival rate to the buffer product, process control computer in accordance with the original in (4) below the arrival rate to the buffer, a step of determining a number of apparatuses for processing each kind dedicated, n _i +1> Λ _i / μ _i > n _i (4) (where i = A, B, the number of devices of the i-th product type is n _i , the arrival rate of the i-th product to the buffer is Λ _i , The service rate in the process is μ _i .)

The process control computer obtains the arrival rate λ _i of the i-th product to the apparatus that commonly processes a large number of products according to the equation (5), and λ _i = Λ _i mod μ _i (5) process The management computer obtains the solutions of the equations (16) and (17) under the conditions of the equations (14) and (15),

[0019]

[Equation 5]

(However, ρ _A and ρ _B are traffic intensities to a device that commonly processes both A and B products of A and B products, and T _{P, A} and T _{P, B} are shared devices. One continuous period for processing products of A type and B type, t _A and t _B are grace (idle) time after the end of the processing of the products of A type and B type in the shared device, and T ₀ is
NB _{, A, limit} , NB _{, B.limit, which} is the sum of product switching time (setup time) in the production cycle, is the upper limit staying amount in the buffer for products of A product and B product. ) The process control computer calculates the allocation cycle T _c of each product by the formula (21), and T _C = K (δ, γ) + δ + γ + T ₀ (21)

(However, K (δ, γ) = T _{P, A} + T _{P, B} , δ = t _A -t _B ,
γ = 2t _B. ) The process control computer is the number of devices n _i required for the process.
And the step of storing the data in the internal memory, the process management computer, the operator required rate R which is the work amount of the operator required for the setup work per unit time accompanying the setup of each device and / or the device Utilization rate U is (2
2) obtaining the expression and / or _{(23), R = 2 / T c (} 22) U = K (δ, γ) / T C ( method of design and operation multi-product production system comprising a 23), A computer-readable recording medium recording the processing program and the processing program is provided.

According to the fifth solution of the present invention, a buffer for each product type that has arrived, an in-process carrier for transporting the product to a predetermined device according to an instruction from the process control computer, a request from an operator, or a priority. And the process control computer that obtains the weighing information of each product, calculates the processing priority of each product, sends a command to the in-process carrier, and assigns it to each device.The process management computer is set in advance. Inputting the input data, the process management computer calculates the arrival rate of the product to the buffer to the corresponding process based on the production plan of each product, and the process management computer receives the data to the buffer. Based on the ratio, according to the equation (4), a step of determining the number of devices dedicated to processing each product type, and n _i +1> Λ _i / μ _i > n _i (4) (where i = A, B, the i-th varieties of the device number n _i, the i-th Arrival rate lambda _i to the buffer varieties, and service rate mu _i at that step.) Step management computer (5) in accordance with formula, to apparatus for shared processing the various kinds of the i-th varieties The step of obtaining the arrival rate λ _i , and λ _i = Λ _i mod μ _i (5) The process control computer obtains the solutions of the equations (16) and (17) under the conditions of the equations (14) and (15). Steps,

[0023]

[Equation 6]

(However, ρ _A and ρ _B are traffic intensities to a device that commonly processes both A and B products of A and B products, and T _{P, A} and T _{P, B} are shared devices. One continuous period for processing products of A type and B type, t _A and t _B are grace (idle) time after the end of the processing of the products of A type and B type in the shared device, and T ₀ is
NB _{, A, limit} , NB _{, B.limit, which} is the sum of product switching time (setup time) in the production cycle, is the upper limit staying amount in the buffer for products of A product and B product. ) The process control computer
The step of calculating the allocation cycle T _c of each product according to the equation (21) and T _C = K (δ, γ) + δ + γ + T ₀ (21) (where K (δ, γ) = T _{P, A} + T _{P, B} , δ = t _A -t _B , γ = 2t _B. ) The process control computer is the number of devices n _i required for the process.
And the step of storing the data in the internal memory, the process management computer, the operator required rate R which is the work amount of the operator required for the setup work per unit time accompanying the setup of each device and / or the device Utilization rate U is (2
2) obtaining the expression and / or _{(23), R = 2 / T c (} 22) U = K (δ, to provide a multi-product production system comprising gamma) / T _C (23).

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings. The present invention is configured by the following embodiments. As described above, as an example, in the production system, the “stay time” of each process is the “waiting time”.
And "service hours". In order to comply with the delivery date, it is necessary to maintain the stay time of each process within a predetermined allowable time.

In the present invention, first, as a first embodiment, a management method for maintaining the process stay time within this predetermined allowable time is disclosed. Then, as a second embodiment,
Disclosed is a personnel planning method and a facility planning method that take into account the work of switching types when a plurality of types are processed. In addition, as a third embodiment, a method of performing product type switching by the equipment and personnel will be disclosed. That is, the priority order of the products to be processed at each time is determined according to the arrival status of each product to the process, and the method of allocating the products according to the order is disclosed.
Furthermore, a specific configuration example of a production system using the above method and its operation will be disclosed. Hereinafter, these embodiments will be described while specifically explaining the present invention.

1. First Embodiment (Management Method for Maintaining Process Stay Time within a Predetermined Allowable Time) The first embodiment shows a case where a product is processed in a certain process. FIG. 1 shows a schematic explanatory diagram of the system.
Here, the product is not a product that has been processed in all steps,
It refers to an object to be processed or a work-in-progress in the process. Let Λ be the arrival rate of this product to the buffer (arrival rate if there is only one device), and let T be the service time in that process.
At this time, the service rate μ is given by μ = 1 / T. When the number of devices is set to n + 1, it is necessary to satisfy Λ · T / (n + 1) <1 (1) in order to perform stable processing, that is, the product does not stay in that process. That is, from a long-term perspective, when the number of devices is n, the product stays in the process and the work in progress continues to increase, and when the number of devices is n + 1, the work in progress can be eliminated. The value of the integer value n can be defined. At present, there are an input buffer for accumulating and placing products arriving at each process, and a warehouse for work-in-process.

The buffer capacity is predetermined for each product. The amount of the product existing in the buffer can be measured, for example, by placing the buffer on a scale and measuring the weight of the product when it enters, or by detecting the light-shielding condition when the product enters by an optical sensor. The in-process carrier manages information on the switching status. When the priority of the product is determined, the computer issues a transportation instruction to the in-process transportation machine. The processing status is reported from the device. The in-process carrier conveys the product, and if it is delivered to the device, control is transferred to the device. In the memory, for example, the information necessary at the time of calculating the priority, that is, the product name, the current number of buffers, the number of arrivals in the period from the previous calculation to the current calculation, the processing time of the product in the device ( This, of course, must be described in advance in the process chart.) In addition, priority calculation time information and the like are stored.

FIG. 2 shows arrival and shipment of products in the process.
It is a figure which shows a relationship. Staying in the buffer at a certain time
The quantity of products and work in progress is the difference between the arrival quantity and the shipment quantity.
Now, period, T_cExplain arrival and shipment between. The number of devices
For n units, T _wDuring the period, the product will accumulate in the buffer
If the number of devices is n + 1, T_pFor the duration of the
The amount of products in it will decrease. At this time, the product in the buffer
Average waiting time W to stay_avExists in the buffer
The average value of the quantity of products is N_avThen, W_av = N_av/ Λ (2) Given in. This is the
This is the relationship known as the formula. As shown in FIG.
When switching the number of units,
Maximum value is N_maxThen, this Little formula is W_av = N_max/ 2Λ (3) Can be rewritten as N_maxIs the number of devices n
It is the value of the in-process amount at the time of switching to n + 1 units. This
The expression of is equivalent to the area of the triangle in the figure, and
It has been multiplied by / 2.

It should be noted that the stay upper limit value in the buffer is N _{B, limit
Assuming} that N _{B, limit} ≧ N _max , the following equation is given as in the equation (3). W _limit = N _{B, limit} / 2Λ (3 ′) These expressions are expressions newly disclosed in the present invention and have an important meaning in managing the production period in the production process.
That is, this formula means that the average value of the waiting time to stay in the buffer can be controlled within W _av time by managing the in-process amount of the product existing in the buffer to be N _max or less. There is. The residence time of the product in each process is the sum of this waiting time and service time as described above. Since the service time is an unchangeable time, the staying time of each process is adjusted by controlling this waiting time.

That is, each process is provided with as many input buffers as there are products produced on the line. According to the first embodiment of the present invention, the permissible quantity corresponding to the permissible value of the average waiting time in the buffer is determined for each type according to the formula (3), and the product flow is controlled so as not to exceed the quantity. It is a form.

2. Second embodiment (personnel planning method in consideration of product type switching work when a plurality of product types are processed,
Facility planning method) The method of calculating the number of devices and the number of personnel in each process on the premise of semi-automatic processing has already been disclosed in the related document (1) formula, and therefore will be omitted here. Here, in the present invention, a method of designing the number of personnel added in accordance with the newly disclosed setup work will be described.

Consider a process in which a plurality of types of products arrive at a certain process. The arrival rate of the i-th variety at the buffer is Λ
i, when the service rate in the process is μ _i , the number of devices n _i that satisfies n _i +1> Λ _i / μ _i > n _i (4) can be determined. Here, the total arrival rate is the number of arrivals of the product to the process per unit time calculated from the production plan of each product. Also,
The service time is the time when the device is exclusively used for processing one processing unit of the product in the process. The n _i devices have
It is assigned to process the i-th product exclusively.
Devices are allocated according to this relational expression for all types of products that arrive at the process.

Next, consider allocation of the remaining products that have not been allocated to the n _i devices. The (n _i +1) th device will process the remaining products of the product type that could not be processed by the n _i device, but that device has a margin (here If there is room,
It has more capacity to process it than it does, which means that there is no more to be processed and processing equipment has more free time. This idle time causes a decrease in the utilization rate of the device. ). Therefore, if the surplus can be sent to the production of other varieties, the utilization rate of the device will be improved.

FIG. 3 is a schematic explanatory view of the allocation of products to the apparatus in the multi-product production. Here, when there are remaining products that cannot be processed by the dedicated device for each product type, instead of allocating equipment to each one, use the same device by sharing it as shown in the figure. By alternately processing multiple types of products, the idle time of the device is reduced and the operating rate is increased. Further, increasing the utilization rate of each device leads to reducing the number of devices.

Regarding the arrival of the i-th product type to the apparatus that commonly processes multiple products, considering again λ _i = Λ _i mod μ _i (5), each product corresponding to this arrival rate is the other product. It will be processed by the same device. Note that λ _i is Λ _i
Represents the remainder when is divided by μ _i . In the following discussion, the arrival rate λ _i calculated by the equation (5) is simply treated as the “arrival rate”. As shown in the figure, the product arrives at the buffer with Λ _A of product A and is allocated to the device dedicated to product A, and the remainder λ _A
Are assigned to the shared device. The same applies to the other varieties B and C. Note that Λ _i is called the “arrival rate to the buffer” of the i-th product type. The problem to be designed at this time is how the setup work is performed according to the combination.

Now, a periodic allocation method according to a second embodiment of the present invention will be disclosed. The basic idea of the cyclic allocation method will be explained. As an example of the premise, consider a case where two or more product types are allocated to the shared device from the situation where each product type is stored in the buffer. First, the parameters will be described. Note that what is shown as "input data" is data that can be set in advance. τ represents the processing time in a certain device. λ represents the arrival amount per unit time to the device or the buffer and the production amount per unit time. μ _i is μ _i ≡1 / τ _i , and represents the service rate when processing i types of products. Λ _i represents the arrival rate of products of i types to the buffer. λ _i represents the arrival rate at the device that commonly processes i-type products together with a plurality of products. N
_{B, i.limit} is _{N B, i.limit ≡2Λ i W av} , i, limit, i
Maximum amount of stay in the buffer of the product of the product type (TAT in the relevant process
(Depending on the total time required for manufacturing, turnaround time). W _{av, i, limit} is the average length of time that i products can stay in the buffer of the process (TAT of the process).
Depends on. This may be used as input data). ρ
_i is ρ _i ≡λ _i / μ _i , and represents the traffic intensity to the device that commonly processes i-type products together with a plurality of types. T
_{P, i} represents one continuous period in which the i-type product is processed in the shared device. T _{W, i} represents a continuous period in which i types of products wait for processing in the shared device in the buffer. t _i is a postponement (idle) after the end of processing on the shared device for i types of products
Represents time. T _c is a production cycle and represents a period from the start of processing of a product of a certain type to the start of the next processing.

First, a certain device (shared device) processes the first product type. When the buffer becomes empty, the processing of that kind is interrupted and the second kind is processed. The first variety that arrives after that will be accumulated in the buffer. When the buffer of the second type is empty, the third type is processed. While the third product is being processed, the first product will continue to be stored in the buffer, and the second product will be newly stored in the buffer. That is, a plurality of other products are processed during the period in which the first product is accumulated in the buffer. When the amount accumulated in the buffer reaches N _max given by the equation (3), the processing of the first type is restarted. If the other varieties before the first varieties reaches N _{m ax} reaches N _max of its varieties, may handle its varieties as the first breed.

It is assumed that a plurality of types of products are assigned while the first product is accumulated in the buffer, and m types of products are allocated during the period T _c in which the first product is processed. Every time the product type is changed, setup is required, requiring operator work. Therefore, the operator request rate, R, is given by R = m / T _c (6). If the time required for setup is known in advance, the number of personnel required for this setup work can be estimated by the theory of queuing.

The method of obtaining T _c will be described below. T _c is represented by T _c = T _{p, i} + T _{w, i} + t _i (7). Where the time T _{p, i} required to process product _i ,
The processing waiting time T _{w, i} and the idle time t _i . Now, assume that there are k kinds of products and that the j-th work-in-progress is waiting in the time buffer until it is processed. At this time, the other types are assumed to be processed in order.

[0041]

[Equation 7]

T ₀ is

[0043]

[Equation 8]

The total time required to exchange k types of products is represented by Applying the relationship of expression (7) to expression (8), and further considering the relationship of expression (3),

[0045]

[Equation 9]

To obtain Basically, if this solution is obtained, the processing time of each product and the processing standby time can be calculated by the equation (8). However, equations (10) and (11) are k-dimensional simultaneous inequalities, and it is difficult to obtain a general solution. Therefore, the solution will now be described by taking as an example the case where two types of products A and B are processed by the same device. When assigning two types to one device, _Tc can be _calculated relatively easily.

FIG. 4 is a diagram showing changes in the product quantity in the buffer when two types are assigned to one unit. Here, the parameters will be described. These parameters are
In particular, in the case of processing two types of products that are developed below, calculation of data required for personnel planning and equipment planning considering the type switching work (used in the calculation of equations (12) to (23)) I will explain. Note that what is shown as "input data" is data that can be set in advance. "Final result data" is obtained based on the input data.

Μ _A (input data) is μ _A ≡1 / τ _A and represents the service rate when processing products of A type. μ _B
(Input data) is μ _B ≡1 / τ _B , and represents the service rate when processing products of B type. Λ _A (input data)
Represents the arrival rate of the product of A type to the buffer. Λ _B (input data) represents the arrival rate of the product of B type to the buffer. λ _A is λ _A ≡ Λ _A mod μ _A.
Shows the arrival rate at the equipment that commonly processes both A and B products. λ
_B is λ _B ≡ Λ _B mod μ _B , and represents the arrival rate of the product of the B type to the device that commonly processes both A and B types. N
_{B, A, limit} (input data) is N _{B, A, limit} ≡ 2Λ _A W
_{av, A, limit,} which indicates the upper limit amount of stay in the buffer for products of the A type (determined by the TAT in the process). N
_{B, B.limit} (input data) is N _{B, B.limit} ≡ 2Λ _B W
_{av, B, limit} , which represents the upper limit stay amount in the buffer of the product of the B type (determined by the TAT in the process). W
_{av, A, limit} represent the average period during which the product of A type can stay in the buffer of the process (determined by the TAT of the process, which may be used as input data). W _{av, B, limit} represents an average period during which the product of the B type can stay in the buffer of the process (determined by the TAT of the process, which may be used as input data). T ₀ (input data) is the total of product type switching time (setup time) during the production cycle. In the case of products of two types A and B, the sum of the setup time from the A type to the B type and the setup time from the B type to the A type is shown. ρ _A is ρ _A ≡ λ _A / μ _A , and represents the traffic strength to the device that commonly processes both A and B products of A product. ρ _B is ρ _B ≡λ _B / μ _B , and represents the traffic intensity to the device that commonly processes both A and B types of B type products. T _{P, A} represents one continuous period (obtained by the solution of FIG. 5) in which the product of A type is processed in the shared device. T _{P, B} represents one continuous period (obtained by the solution of FIG. 5) in which the B type product is processed in the shared device.
T _{W, A} represents a continuous period in which the product of A type is waiting for processing in the shared device in the buffer. T _{W, B} represents a continuous period in which the product of the B type is waiting for processing in the shared device in the buffer. t _A represents the grace (idle) time after the end of processing in the shared device of the product of A type. t _B represents the grace (idle) time after the end of the processing of the B type product in the shared device.
δ is δ≡t _A -t _B , and represents a parameter selected by the user of this system. γ is γ≡2 t _B and represents a parameter selected by the user of this system.

By properly selecting these parameters δ and γ,
The operator request rate and equipment utilization rate of the final result will be determined. T _C is T _C ≡K (δ, γ) + δ + γ + T ₀ ,
Production cycle. Equal for both types of products. It represents the period from the start of processing to the start of the next processing. K (δ, γ) is K (δ, γ)
≡ T _{P, A} + T _{P, B} , and represents the sum of the processing period of the A type product and the B type product.

The final result data R is R≡2 / T _C , and the operator request rate represents the number of times the operator is required for the setup work per unit time. When switching between two types, setup work is required twice during the period of Tc. The final result data U is, U≡K (, δ, γ) are / T _C, represents the utilization of the device. The formula is expanded below. Since two types are assumed here, equations (10) and (11) are as follows.

[0051]

[Equation 10]

It is Based on FIG. 4, equation (12), (1
Let us further consider the equation 3). Now, let t _{A be} the idle time after processing the A product, and t _{B be} the idle time after processing the B product. (This is a parameter for absorbing variations in processing time and arrival interval.) By introducing the idle time, the inequalities of the equations (12) and (13) become the following equations.

[0053]

[Equation 11]

To obtain. Further, from FIG. 4, t _A + t _B = T _C- (T _{P, A} + T _{P, B} ) -T _O ≡δ + γ (19). δ is a parameter that indicates the difference between the production cycles of the two types, and γ is a parameter that depends on the production cycle. These are, so to speak, leisure time. The period for processing the A type and the period for processing the B type are obtained, and T _c can be obtained from the result.

Since N _{A, Max} is smaller than N _{B, A, limit} , the delivery date is satisfied. That is, the setup plan and the delivery date are determined by managing the amount of products in the buffer.
The delivery date here is the delivery date of each process, and is the sum of the product staying time in the buffer and the product processing time in the apparatus. It is not possible to set a delivery date that is shorter than the processing time. The time spent in the buffer is given by equation (3),
The processing time is a time peculiar to the process and is described in advance in the process chart.

Next, FIG. 5 shows an explanatory diagram of how to obtain a solution when allocating two kinds of products. This figure shows (16),
It shows a solution under the equations (14) and (15) of the equation (17). After all, T _{P, A} and T _{P, B} can be optimized using δ and γ as parameters. A solution can be obtained from K _min to K _max , but K _min is “0” for both δ and γ.
It corresponds to the case of. Kmax depends on the formula (14) or the formula (15), and the maximum value of the work-in-process that can stay in the buffer is restricted by the demand of the production period. Here, K (δ, γ) is defined as K (δ, γ) = T _{P, A} + T _{P, B} (20). Therefore, K (δ, γ) is δ and γ
Can be optimized as a parameter.

The calculator is within the range from K _min to K _max.
Find the solution of K. That is, while increasing δ and γ,
Repeat the calculation. When K is obtained, (20) (2
The operator request rate and the operation rate are obtained from the equations (1), (22), and (23).

Next, a method of selecting a product type to be processed by the same device will be described. First, we examine whether or not it is possible to combine products with different arrival rates to minimize the personnel request rate and maximize the device utilization rate. Furthermore, since K (δ, γ) is determined as shown in Fig. 5, we will examine how the personnel demand rate and the equipment utilization rate change depending on δ and γ, and select the combination of product types according to what policy, Consider whether δ and γ should be chosen.

FIG. 6 is a diagram showing the relationship between traffic intensity and utilization rate when processing two types of products. In the figure, the vertical axis represents the utilization factor of the device, the horizontal axis represents the traffic intensity of one type, and the traffic intensity of the other type is used as a parameter. Further, here, the allowable residence time in the buffer and δ and γ are set to constant values. When a B type with a certain traffic strength is defined, the device utilization rate increases as the traffic strength of the other type increases. At this time, there is a certain maximum value in the device utilization rate. This maximum value slightly increases as the traffic intensity of the B product increases, and becomes maximum when the traffic intensity of the B product and the traffic intensity of the A product are equal. Furthermore, as the traffic intensity of B type increases, the device utilization rate decreases. Further, the limit value of the device utilization rate becomes lower as the setup time becomes longer. As a result, T _c is expressed by the equation (21) as a two-variable function of δ and γ. This relationship can also be obtained from FIG. T _C = K (δ, γ) + δ + γ + T ₀ (21) It was possible to obtain T _c when processing two types of products with one device. The operator request rate R is R = 2 / T _c (22) Further, the average utilization rate U is U = K (, δ, γ) / T _C (23) for the switched n _i + 1st device. Is.

As a method of combining the two types, it is possible to combine a type with a high arrival rate and a type with a small arrival rate. When this method is applied, it can be easily inferred that the operating rate of each device becomes substantially uniform. However, when the operation rate is large, the equations (16) and (17) are (14)
There may be no solution under the conditions of the equation and the equation (15).
This situation is shown in FIG. The range with solutions narrows as the setup time increases. For example, when the setup time is 20 unit hours, there is no type that can be combined with a type whose arrival rate exceeds about 0.83. In this case, a dedicated device is assigned to the product having a high arrival rate. Since the number of devices and the number of personnel calculated here are limited to two types of products that can be assigned to one device, it is not the minimum in concrete operation described later, but at least a value that does not cause a shortage in operation. You can say that. That is, the required upper limit value of the equipment and the number of personnel in the production system design is given.

3. Multi-product production system and calculation process of equipment / number of personnel (specific configuration of production system enabling management of the present invention) As described above, in the present invention, an input buffer is installed in each product and A system having a function of monitoring the in-process amount of existing products and notifying the operator when the in-process amount reaches a certain value will be configured. At this time, the production system instructs a new priority order of products to be processed. The operator who receives the notification may stop the processing of the product having the lower priority, perform the setup work for the product having the higher priority, and then switch the transport route from the buffer to the apparatus.

FIG. 7 shows a block diagram of a multi-product production system. The operation of the system will be described below according to this specific example. This system includes a line management computer 1, a process management computer 2, a distributor 3, a buffer 4, an intra-process carrier 5, an apparatus 6, an aggregator 7, and a line management network 8. The line management computer 1 and the process management computer 2 each have a memory for storing necessary data therein. Further, both computers send and receive various data such as allowable stay time and progress status information through the line management network 8.

The line management computer 1 determines the allowable stay time in each process based on the delivery date information of each product and notifies the process management computer 2 of the allowable stay time. Further, the line management computer 1 grasps the new production progress status of each product and changes the allowable stay time in each process as necessary. The process control computer 2 specifically controls the processing of each product. Each product first reaches the distributor 3. The distributor 3 identifies the product type and sends it to the input buffer 4 for each product. This process can be automated by the process control computer if the identification code for each product is attached.

When the processing speed of the distributor 3 is slow, it is necessary to install a buffer in front of the distributor 3. The input buffer 4 of each type can be automated by providing a weighing sensor or the like. The weighing function can be realized simply by measuring the weight. There is also a method of using various optical sensors. The process control computer 2 obtains the weighing information of each product, calculates the processing priority of each product, and sends a command to the in-process carrier 5. The in-process carrier machine 5 follows the instructions of the process management computer 2, the request from the operator, or the priority,
Transport the product to the intended equipment. At present, the setup work is carried out manually, so the designation of the device for the in-process carrier machine 5 may be carried out manually, or automatically by the line management computer 1 or the process management computer 2. good. If setup work is required,
The process control computer 2 emits a signal and requests operator assistance. Stop some devices as shown in "(Method of allocating products according to the order of priority of products to be processed at each time according to the arrival status of each product to the process") It is also possible to notify the operator of the determination to be made based on the weighing information from each input buffer 4.

Next, FIG. 8 shows a flowchart for calculating the number of facilities and personnel in the multi-product settlement system. Here, the procedure for calculating the number of equipments and the number of personnel in consideration of the product type switching work at the time of multi-product production will be summarized again.
First, predetermined input data set in advance is set in the internal memory of the line management computer 1 or the process management computer 2. The process management computer 2 reads various kinds of predetermined input data from the internal memory or the line management computer 1 (S101). The process control computer 2 calculates the number of personnel and service time required for the pre-processing and the post-processing by a known method (S102). This calculation method is disclosed, for example, in Related Document (1). Next, the process management computer 2 calculates the total arrival rate of products to the process based on the production plan of each product (S103). Based on the total arrival rate, the process control computer 2 determines the number of devices dedicated to processing each product according to the equation (4) (S105). Next, the process control computer 2 obtains the arrival rate according to the equation (5) (S107). Further, the process management computer 2 combines the types with high arrival rate and the types with small arrival rate, and (14)
Under the conditions of equations (15) and (15), solutions of equations (16) and (17) are obtained (S109). If a solution cannot be obtained, a dedicated processing device is assigned to a product having a high arrival rate. Next, the process control computer 2 calculates the allocation cycle of each product by the formula (21) (S111).

The process control computer 2 executes steps S105-S5.
The number of devices required for the process can be determined by 111, and the process management computer 2 stores the data in the internal memory (S113). In addition, the process control computer 2
As a result of performing the calculations in steps S107 to 111, the operator request rate associated with the setup of each device is calculated, and the operator request rate for the setup work of the process is obtained by the formula (22) as a sum thereof (S115). ). further,
The process control computer 2 obtains the utilization factor of the apparatus by the equation (23). The process control computer 2 stores the obtained operator request rate and / or device utilization rate in a memory. The process management computer 2 can determine the number of personnel required for the setup work based on the operator request rate obtained in step S115, and the process management computer 2 stores the data in the internal memory (S117). The process control computer 2 is step S
The total number of personnel required for the production system is determined by adding the number of personnel required for the pre-processing and the post-processing obtained in 101, and the data is stored in the internal memory (S119). Further, the process control computer 2 may store the respective intermediate result data and / or final result data in the internal memory and / or transmit them to the line management computer 1 each time they are obtained or as appropriate. Line management computer 1
Stores data obtained as appropriate in the internal memory.

As described above, in the multi-product production system, based on the production plan of each product, the second aspect of the present invention concerning the equipment plan and the personnel plan in consideration of the decrease of the operation rate due to the product switching work. Embodiments have been disclosed. 4. Third embodiment (a method of allocating varieties according to the order of priority of varieties to be processed at each point in time according to the arrival status of each variety to the process)

The above-described method "(personnel planning method and equipment planning method in consideration of product switching work when a plurality of products are processed)" is merely a designing method for a multi-product production system. That is, it is premised that the production plan is determined and the product flow proceeds according to the plan. In the actual production system, the product flow does not proceed as planned.
For this reason, it is necessary to switch the type assigned to the device in real time. Here, the switching method is disclosed.

As already described in “(Management method for maintaining process staying time within predetermined allowable time)”, by setting the in-process amount of the product staying in the input buffer to be N _max or less, The average stay time can be maintained within the allowable stay time. Now, consider that multiple types of products are processed in a process. It is assumed that the number of devices is prepared by the method described in “(Personnel planning method and equipment planning method that considers product switching work when multiple products are processed)”.

Now, as an example, consider a case where seven types of products are processed by three devices. The situation in which those products are retained in the input buffer shall be observed every unit time. As described in “(Management method for maintaining process stay time within a predetermined allowable time)”, the buffer is prepared for each product type. Let the retention quantity of the i-th variety be N _i, and the allowable quantity of this variety be N _{i, max} . If the priority factor that determines the allocation order is defined by P _i = N _i / N _{i, max} (24), if the product with the largest P _i value is allocated,
It is possible to comply with the allowable residence time while keeping the operating rate of the device high.

The arrival rate of each product is not taken into consideration in the priority calculated by the equation (24). If the arrival rate of a certain product is high, more than the allowable quantity of products will remain in the buffer as work in process by the next observation point. For this reason, it is possible to estimate the arrival amount until the next observation time and determine the priority in consideration of it. The priority factor in this case is as shown in equation (25). The method of estimating the arrival rate until the next observation point will be described later. P _i = {N _i + Λ _i T} / N _{i, max} (25)

Λ _i is the arrival rate of the i-th product type to the buffer. T is the observation period of the buffer. At this time, the arrival rate of this kind to the buffer is large, and according to the equation (4),
When n _i is 1 or more, the relevant product is assigned to the device as described in (Personnel planning method and facility planning method in consideration of product switching operation when multiple products are processed).

The priority shown by the equation (25) does not take into consideration the reduction of products in the buffer due to the processing. The product in-process volume in the buffer decreases at a rate of μ _i −λ _i per unit time. μ _i is the service rate of the product and the reciprocal of the service time. The expression (5) means that μi is larger than λi, which means that the in-process amount of the product in the buffer always decreases in the long term. The time at which the product in-process in the buffer is cleared, t, is t = Ni / (μ _i −λ _i ). The longer this time is, the more susceptible the production system is to changes in the operating conditions of the production equipment. Therefore, preferentially allocating to a device from a product with a long time is less affected by operational fluctuations. In order to consider the difference in the allowable buffer quantity for each product type, the priority factor is standardized by N _{i, max} ,
Equation 6) is obtained. P _i = N _i / {N _{i, max}・ (N _{u, i}・ μ _i ―Λ _i )} (26)

Here, the traffic intensity ρ _i = Λ _i of the product _i
The value including the decimal point of / μ _i is N _{u, i} . When the change in the arrival rate is small, the priority can be determined also by the expression (27). P _i = N _i / {N _{i, max} · (μ _i ―λ _i )} (27) In equations (25) and (26), Λ _i and λ _i are future values based on past performance values. This is an estimated value that is expected to arrive. Several methods are known as methods for estimating future values based on past actual values (related material (4): NRSand
ers, 'Forecasting Guidelines and Methods', in
'ENCYCLOPEDIA OF PRODUCTION AND MANUFACTURING MAN
AGEMENT 'edited by Paul M. Swamidass, Kluwer Acad
emic., pp.228-235 (2000)). In the embodiment of the present invention, a method called Single Exponential Smoothing (SES) method (single exponential smoothing method) is adopted.

FIG. 9 is a diagram showing the result of the simulation. Consider a process in which 7 types of products are processed by 3 units. As mentioned earlier, in this case,
Since it is generally difficult to obtain a solution, we performed a solution by simulation. Although the production plan for the 7 types is fixed, in other words, the average arrival rate to the process is fixed, but it is assumed that there is fluctuation in arrival due to fluctuations in the operating conditions. Here, the arrival that follows the Poisson distribution in queuing theory is generated by random numbers and is simulated. The in-process amount of the product in the buffer is to be observed every 100 unit hours. For the observation cycle, it is possible to select the smallest value of T _c obtained by the method described in (Personnel planning method and equipment planning method that considers product switching work when multiple products are processed). To be In this simulation, the priority of product allocation to the device is determined by the equation (27).

In this example, the simulation is started from the state where the initial work-in-process exists in advance. Over time,
It can be seen that the total amount of products in process in the buffer is decreasing. At this time, the operator has 1
At every 00 unit time, the maximum number of people corresponding to the number of devices is required. However, the observation cycle can be obtained by the method described in "(Personnel planning method and equipment planning method that considers product switching work when multiple products are processed)"
Since it is set as the smallest value among T _c , "(personnel planning method and equipment planning method that considers product switching work when multiple products are processed)"
There is no shortage of personnel required in. If there is sufficient product work in each buffer, the equipment availability is 100.
It was confirmed by simulation that a value close to% could be maintained. When the amount of work in process of each device is small and all the work in process is completed within the observation period, the operation rate of the device decreases.

The following management is performed in order to keep the operating rate high. For example, the total sum of all in-process quantities of each product type is monitored, and when the value falls below a certain value, the number of devices is reduced. The in-process amount of the product in the buffer increases, but as long as this value does not exceed N _{i, max} , the product is processed within the allowable stay time in the process. By appropriately selecting the cycle of changing the number of devices, it is possible to allocate the period in which the number of devices is reduced to maintenance work such as inspecting the suspension device, which is also a good effect on the stability of the production system.

The multi-product production system design / operation method of the present invention comprises a multi-product production design / operation method processing program for causing a computer to execute each procedure, and a multi-product production system design / operation method processing program. It can be provided by a recorded computer-readable recording medium, a program product including a design / operation method processing program thereof and loadable in an internal memory of a computer, a computer such as a server including the program, and the like.

[0079]

The present invention has disclosed a facility plan and a personnel plan in consideration of the setup work associated with the product type switching in a multi-product production line. When the method disclosed heretofore is applied to a multi-product production system, the amount of machinery and equipment and the number of personnel are designed to be too small, and when production is performed with such equipment and personnel, there is a shortage of equipment and personnel. However, according to the present invention, a highly accurate design method in a multi-product production system is clarified, which is effective for further improvement of productivity such as reduction of production cost.

[Brief description of drawings]

FIG. 1 is a schematic explanatory diagram of a system.

FIG. 2 is a diagram showing a relationship between arrival and shipment of products in the process.

FIG. 3 is a schematic explanatory diagram of allocation of products to devices in multi-product production.

FIG. 4 is a diagram showing changes in the product quantity in the buffer when two types are assigned to one unit.

FIG. 5 is an explanatory diagram of how to obtain a solution when assigning two types.

FIG. 6 is a diagram showing the relationship between traffic intensity and usage rate when processing two types of products.

FIG. 7 is a block diagram of a multi-product production system.

FIG. 8 is a flowchart for calculating the number of facilities / personnel in the multi-product settlement system.

FIG. 9 is a diagram showing an implementation result by simulation.

[Explanation of symbols]

1 line management computer 2 process control computer 3 distributors 4 buffers 5 In-process carrier 6 devices 7 Aggregator 8 line management network

Claims (16)

[Claims] 1. A process for processing a product in a multi-product production system is provided with a buffer to be input for each product so that the time that the product stays in the buffer is within an average permissible time during which the product can be stayed. A method for designing and operating a multi-product production system characterized by defining a control value for a buffer. 2. An average waiting time W _av for a product to stay in the buffer when a buffer to be input for each product is provided and the number of devices is switched in a product processing step in a multi-product production system Let N _{max be} the maximum in-process amount to stay and Λ be the arrival rate of the product to the buffer. W _av = N _max / 2Λ (3), and the in-process amount of the product existing in the buffer is N _max or less. A method for designing and operating a multi-product production system that controls the average waiting time to stay in the buffer within _Wav hours by managing the above. 3. In a processing process for processing a plurality of products, which is composed of a plurality of devices, a maximum of two kinds of products are allocated to one device, and an allocation cycle is calculated for the devices to which the two kinds of products are allocated. A method for designing and operating a multi-product production system, which is characterized by designing the number of personnel required for setup work. 4. The process management computer inputs the preset input data, and the process management computer calculates the arrival rate of the product to the corresponding process buffer based on the production plan of each product type. The step of calculating and the process control computer, based on the arrival rate to the buffer (4)
According to the formula, the step of determining the number of devices dedicated to processing each type, and n _i +1> Λ _i / μ _i > n _i (4) (where i = A, B, the number of devices of the i-th type is n _i , the arrival rate of the i-th product type to the buffer is Λ _i , and the service rate in that process is μ _i .) The process management computer is designed to use a device that commonly processes multiple products according to equation (5). Of the i-th product arrival rate λ _i , and λ _i = Λ _i mod μ _i (5) The process control computer sets (16) and (15) under the conditions of equations (14) and (15). 17) The step of obtaining the solution of the equation, and (However, ρ _A and ρ _B are the traffic intensities to the devices that commonly process both A and B products of A and B products, and T _{P, A} and T _{P, B} are _A devices of shared products and One continuous period for processing products of type B, t _A and t _B are grace (idle) time after the end of processing of shared products of type A and products of type B, and T ₀ is the type during the production cycle Switching time (setup time)
, N _{B, A, limit} , N _{B, B.limit} are the upper limit staying amount in the buffer for the products of A type and B type. ) The process control computer calculates the allocation cycle T _c of each product according to equation (21), and T _C = K (δ, γ) + δ + γ + T ₀ (21) (where K (δ, γ) = T _{P , A} + T _{P, B} , δ = t _A −t _B , γ = 2t _B. ) The process control computer is the number of devices n _i required for the process.
And storing the data in the internal memory, the process control computer, the operator demand rate R which is the work amount of the operator required for the setup work per unit time accompanying the setup of each device and / or the device Utilization rate U, (22)
A method for designing and operating a multi-product production system including the step obtained by the equation and / or the equation (23) and R = 2 / T _c (22) U = K (δ, γ) / T _C (23). 5. The input step comprises the service rates μ _A and μ _B when processing products of the A and B types, arrival rates Λ _A and Λ _B of the products of the A and B types to the buffer, Sum of product switching time (setup time) during the production cycle
The maximum amount of stay in the buffer is N for T ₀ , A and B products
_The method for designing and operating a multi-product production system according to claim 4, wherein _{B, A, limit} and N _{B, B. limit} are input. 6. The multi-product production system according to claim 4, wherein the process control computer further comprises a step of calculating the number of personnel and service time required for the pre-processing and / or the post-processing. Design and operation method. 7. The multi-product according to claim 4, further comprising the step of assigning a device for exclusive use to a product having a high arrival rate when the solution is not obtained in the step of obtaining the solution. Production system design and operation method. 8. An arrival rate estimated from past results by periodically monitoring the quantity of products in a buffer in a processing process including a plurality of devices and an input buffer provided for each product type. , The service rate of processing the product,
A method for designing and operating a multi-product production system, characterized in that the management value of the buffer is used to determine the priority of the product to be processed and allocate it to the device. 9. A processing step comprising a plurality of devices and an input buffer provided for each product type, the quantity of products in the buffer is periodically monitored, and the product is retained in the i-th product type buffer. Let the quantity be N _i ,
When the allowable quantity that can be retained in the buffer when processing this product within the time defined in the process is N _{i, max} ,
A method for designing and operating a multi-product production system, characterized in that a priority factor P _i that determines an allocation order is defined by the following equation, and products having a large value of the priority factor P _i are allocated to a processing device. P _i = N _i / N _{i, max} (24) 10. A process step comprising a plurality of devices and an input buffer provided for each product type, the quantity of products in the buffer is periodically monitored, and the product is retained in the i-th product type buffer. The quantity is N _i , the allowable quantity that can be retained in the buffer when this product is processed within the time defined by the process is N _{i, max} , the arrival rate of the i-th product at the buffer is Λ _i , the buffer The observation period of T,
Then, the priority factor P _i that determines the allocation order is defined by the following formula, and the design / operation method of the multi-product type production system is characterized in that the products having a large value of the priority factor P _i are allocated to the processing devices. P _i = {N _i + Λ _i T} / N _{i, max} (25) 11. A processing step comprising a plurality of devices and an input buffer provided for each product type, the quantity of products in the buffer is periodically monitored, and the product is retained in the i-th product type buffer. quantity of N _i, estimate the varieties expected arrival of future residence allowable quantity N _{i, max,} for the arrival rate of the i-th varieties in the buffer when processing in a time determined in the step Value Λi, i
The service rate when processing a product of a product type is μ _i , and the value of the traffic strength ρ _i = Λ _i / μ _i of product type i including the fractional part is N
_{Let u, i be} the priority factor P _i that determines the allocation order.
Is defined by the following equation, and the design and operation method of the multi-product production system is characterized by allocating to the processing device from the product having the larger value of the priority factor P _i . P _i = N _i / {N _{i, max}・ (N _{u, i} / μ _i ―Λ _i )} (26) 12. In a processing step comprising a plurality of devices and an input buffer provided for each product type, the quantity of products in the buffer is periodically monitored, and the product is retained in the i-th product type buffer. The quantity is N _i , the allowable quantity that can be retained in the buffer when processing this product within the time defined in the process is N _{i, max} , and the service rate when processing products of i product is μ _i , the past The estimated arrival rate at the buffer based on the actual value of
_i, and the time, process control computer on the basis of this, in accordance with the following equation (4), determine the number of apparatuses for processing dedicated to each _{variety, n i +1> Λ i /} μ i> n i ( 4) (However, i = A, B, the number of devices of the i-th product type is n _i , the arrival rate of the i-th product type to the buffer is Λ _i , and the service rate in the process is μ _i .) After assigning n _i devices to the i-th product type, the process control computer assigns the first device to the device that processes multiple products according to equation (5).
The arrival rate λ _i of the i-th variety is calculated, and λ _i = Λ _i mod μ _i (5) The priority factor P _i that determines the allocation order is defined by the following equation, and the absolute value of the priority factor P _i is large. A method for designing and operating a multi-product type production system characterized by allocating from a processor to a processing device. P _i = N _i / {N _{i, max}・ (μ _i ―λ _i )} (27) 13. Obtaining a buffer for each product type that has arrived, an in-process transporter that transports the product to a predetermined device according to an instruction of a process control computer, a request from an operator or a priority, and weighing information of each product. The process control computer that calculates the processing priority of each product, sends a command to the in-process carrier, and assigns each device to the process management computer, and the process management computer inputs the preset input data, The process management computer calculates the arrival rate of the product to the buffer to the corresponding process based on the production plan of each product, and the process management computer calculates the arrival rate to the buffer based on the arrival rate to the buffer (4).
According to the formula, the step of determining the number of devices dedicated to processing each type, and n _i +1> Λ _i / μ _i > n _i (4) (where i = A, B, the number of devices of the i-th type is n _i , the arrival rate of the i-th product type to the buffer is Λ _i , and the service rate in that process is μ _i .) The process management computer is designed to use a device that commonly processes multiple products according to equation (5). Of the i-th product arrival rate λ _i , and λ _i = Λ _i mod μ _i (5) The process control computer sets (16) and (15) under the conditions of equations (14) and (15). 17) The step of obtaining the solution of the equation, and (However, ρ _A and ρ _B are the traffic intensities to the devices that commonly process both A and B products of A and B products, and T _{P, A} and T _{P, B} are _A devices of shared products and One continuous period for processing products of type B, t _A and t _B are grace (idle) time after the end of processing of shared products of type A and products of type B, and T ₀ is the type during the production cycle Switching time (setup time)
, N _{B, A, limit} , N _{B, B.limit} are the upper limit staying amount in the buffer for the products of A type and B type. ) The process control computer calculates the allocation cycle T _c of each product according to equation (21), and T _C = K (δ, γ) + δ + γ + T ₀ (21) (where K (δ, γ) = T _{P , A} + T _{P, B} , δ = t _A −t _B , γ = 2t _B. ) The process control computer is the number of devices n _i required for the process.
And storing the data in the internal memory, the process control computer, the operator demand rate R which is the work amount of the operator required for the setup work per unit time accompanying the setup of each device and / or the device Utilization rate U, (22)
A multi-product production system including steps obtained by the formula and / or formula (23), and R = 2 / T _c (22) U = K (δ, γ) / T _C (23). 14. The allowable stay time in each process is set based on the delivery date information of each product and notified to the process management computer, the production progress status of each product is grasped, and allowed in each process as necessary. The multi-product production system according to claim 1, further comprising a line management computer for changing a stay time. 15. A step of inputting preset input data by a computer, and a step of calculating the arrival rate of a product at a buffer to a corresponding process based on a production plan of each product type. The computer determines, based on the arrival rate to the buffer, the number of devices dedicated to processing each product according to equation (4), and n _i +1> Λ _i / μ _i > n _i (4) ( However, i = A, B, the number of devices of the i-th product type is n _i , the arrival rate of the i-th product type to the buffer is Λ _i , and the service rate in that process is μ _i .) According to the equation (5), the step of obtaining the arrival rate λ _i of the i-th product to the apparatus that commonly processes a large number of products, and λ _i = Λ _i mod μ _i (5) The computer executes (14) and (15). ) Equation (16) and (17) under the condition of the equation, ] (However, ρ _A and ρ _B are the traffic intensities to the devices that commonly process both A and B products of A and B products, and T _{P, A} and T _{P, B} are _A devices of shared products and One continuous period for processing products of type B, t _A and t _B are grace (idle) time after the end of processing of shared products of type A and products of type B, and T ₀ is the type during the production cycle Switching time (setup time)
, N _{B, A, limit} , N _{B, B.limit} are the upper limit staying amount in the buffer for the products of A type and B type. ) The computer calculates the allocation cycle T _c of each product according to the equation (21) and T _C = K (δ, γ) + δ + γ + T ₀ (21) (where K (δ, γ) = T _{P, A} + T _{P, B} , δ = t _A −t _B , γ = 2t _B. ) The computer determines the number n _i of devices required for the process and stores the data in the internal memory. The management computer sets the operator request rate R and / or the device utilization rate U, which is the work amount of the operator required for the setup work per unit time accompanying the setup of each device, to (22)
Formula and / or step obtained by formula (23) and design of multi-product production for causing computer to execute R = 2 / T _c (22) U = K (δ, γ) / T _C (23)
Operation processing program. 16. A computer inputs a preset input data, and a computer calculates an arrival rate of a product to a buffer to a corresponding process based on a production plan of each product type. The computer determines, based on the arrival rate to the buffer, the number of devices dedicated to processing each product according to equation (4), and n _i +1> Λ _i / μ _i > n _i (4) ( However, i = A, B, the number of devices of the i-th product type is n _i , the arrival rate of the i-th product type to the buffer is Λ _i , and the service rate in that process is μ _i .) According to the equation (5), the step of obtaining the arrival rate λ _i of the i-th product to the apparatus that commonly processes a large number of products, and λ _i = Λ _i mod μ _i (5) The computer executes (14) and (15). ) Equation (16) and (17) under the condition of obtaining the solution, ] (However, ρ _A and ρ _B are the traffic intensities to the devices that commonly process both A and B products of A and B products, and T _{P, A} and T _{P, B} are _A devices of shared products and One continuous period for processing products of type B, t _A and t _B are grace (idle) time after the end of processing of shared products of type A and products of type B, and T ₀ is the type during the production cycle Switching time (setup time)
, N _{B, A, limit} , N _{B, B.limit} are the upper limit staying amount in the buffer for the products of A type and B type. ) The computer calculates the allocation cycle T _c of each product according to the equation (21) and T _C = K (δ, γ) + δ + γ + T ₀ (21) (where K (δ, γ) = T _{P, A} + T _{P, B} , δ = t _A −t _B , γ = 2t _B. ) The computer determines the number n _i of devices required in the process and stores the data in the internal memory, and the process The management computer sets the operator requirement rate R and / or the equipment utilization rate U, which is the work amount of the operator required for the setup work per unit time accompanying the setup of each device,
(22) obtaining the expression and / or _{(23), R = 2 / T c (} 22) U = K (δ, γ) / T C (23) of multi-product production for causing a computer to execute design·
A computer-readable recording medium recording an operation processing program.