Method and apparatus for improving manufacturing processes

I claim:

1. A method of scheduling and operating production of a factory having a plurality of work stations, each work station performing one or more processes, in order to meet a predetermined shipping schedule, comprising the steps of:

(a) determining the rate of material flow out (FLOWOUT.sub.ij) of each process within each workstation of a factory necessary for the factory to meet a predetermined shipping schedule;

(b) determining the size of the batch of material for each process of each workstation necessary to meet each material flow rate determined in step (a) for the workstation; and

(c) operating each process at each workstation with the batch sizes determined in step (b).

2. The method for scheduling and operating production of a factory as set forth in claim 1 wherein the size of the batch of material for process i at workstation j is determined in step (b) substantially in accordance with the following relationship:

where, for process i within workstation j, B.sub.ij is the batch size, FLOWOUT.sub.ij is the rate of material flow out of the process in the workstation necessary for the factory to meet a predetermined shipping schedule, CT.sub.j is the cycle time of the workstation, %SCRP.sub.ij is the percentage of scrapped material produced by the process and SUSCRP.sub.ij is the amount of scrapped material produced by the set-up of the process.

3. The method for scheduling the operating production of a factory as set forth in claim 2 wherein CT.sub.j is determined based on the based upon the following relationship:

where, for process i within workstation j, SUeff.sub.ij is the effective set-up time, Peff.sub.ij is effective time to process one unit of material, FLOWOUT.sub.ij is the material output flow rate, and n is the number of processes performed by workstation j.

4. The method for scheduling and operating production according to claim 3 further comprising the steps of:

redetermining Peff.sub.ij if the process i produces rework according to,

where RP.sub.ij is the rework processing rate and %REWK.sub.i is the AMOUNT percent of material reworked; and

redetermining Peff.sub.ij if process i within workstation j produces scrap according to,

where %SCRP.sub.ij is the percentage of scrap.

5. The method for scheduling and operation production according to claim 3 further comprising the steps of:

redetermining Peff.sub.ij if process i within workstation j has associated with it human down time according to,

where MTTB.sub.ij equals the mean time between human breaks and MTOB.sub.ij equals the mean time on break; and

redetermining Peff.sub.ij if process i has associated with it machine down time according to,

where MRBF.sub.ij equals the means time between machine failures and MTTR.sub.ij equals the mean time to repair.

6. The method for scheduling and operating production according to claim 3 further comprising the step of:

determining SUeff.sub.ij if process i within workstation j has associated with it rework according to

where RSU.sub.ij is the rework setup time.

7. The method for scheduling and operating production according to claim 3 further comprising the step of:

redetermining SUeff.sub.ij if process i has associated with it human down time according to,

where MTTB.sub.ij is the mean time to break and MTOB.sub.ij is the means time on break.

8. The method for scheduling and operating production according to claim 3 further comprising the step of:

determining the percentage of time that the workstation is processing material.

9. The method for scheduling and operating production according to claim 3 further comprising the step of:

determining the time during which the workstation is down due to human down time and machine down time.

10. A method for scheduling and operating a factory having a plurality of workstations, each workstation performs one or more processes, each process operating on a batch of material, the method comprising the steps of:

determining the rate of material flow out (FLOWOUT.sub.ij) of each process within each workstation of a factory for the factory to meet predetermined shipping rates;

determining the cycle time CT.sub.j for each workstation j to process a batch of material based upon the following relationship:

where, for process i within workstation j, SUeff.sub.ij is the effective set-up time, Peff.sub.ij is effective time to process one unit of material, FLOWOUT.sub.ij is the material output flow rate, and n is the number of processes performed by workstation j; and

identifying inefficient workstations having the longest cycle times by an evaluation of cycle times and batch sizes; and

operating the processes performed by the identified workstations having the longest cycle times to reduce the batch size and cycle time of the identified workstations.

11. The method of claim 10 wherein the cycle time CT.sub.j is determined when the factory is operating at capacity to manufacture a mix products at the shipping rates.

12. The method of claim 11 wherein the step of determining CT.sub.j when the factory is operating at capacity includes the step of linearly adjusting the flow rates such that none of the workstations are producing at overcapacity and one or more workstations are at capacity.

13. A factory for manufacturing and shipping goods, the factory comprising:

a plurality of workstations, each workstation performing one or more processes, each process processing a batch of material having sizes determined by a system for scheduling manufacturing processes to meet a shipment schedule; and

a system for scheduling manufacturing processes to meet a shipment schedule, the system including: a memory for storing data values of variables received from the factory representing shipping rates of a mix of products produced by the factory, processing times per unit of material for each process in each workstation and set-up times for each process in each workstation; a microprocessor, coupled tot he memory, for determining the rate of material flow out of each process within each workstation necessary for the factory to meet the shipment schedule and for determining the size of the batch of material for each process of a workstation necessary to meet each material flow rate out of each workstation for each process; and a means for providing the batch size for a process to the factory.

14. A factory production scheduling and organization process, the process outputting information for optimizing production scheduling and organization and improving the efficiency of a factory production line to meet a predetermined shipping rate, the factory production line comprised of a plurality of interconnected workstations each performing at least one process, comprising the steps of:

receiving factory production line organization and operation data;

modelling the factory production line, according to the input factory data, by generating flow rates of material through each workstation for each process and processing capacities of each workstation for each process to meet the predetermined shipping rate for the production line;

analyzing the factory model according to the factory data and generated workstation material flow rates and workstation processing capacities for each process to evaluate workstation performance, process and production line performance, and identify optimal production line operating and supply characteristics; and

operating the production line to implement the identified optimal production line operating and supply characteristics to substantially meet the predetermined shipping rate.

15. The factory production scheduling and organization process as in claim 14 wherein the step of modelling the production line comprises the steps of:

determining the effective processing time per material unit produced by each workstation for each process;

determining the effective set-up time for reconfiguring the process performed by each workstation;

generating, for each workstation in the production line, the flow rates of material through each workstation necessary to meet the predetermined shipping rate; and

generating, from the material flow rates and effective processing time for each process performed by each workstation, the processing capacity of each workstation for each process.

16. The factory production scheduling and organization process as in claim 15 wherein the step of analyzing the factory model comprises the steps of:

determining material scheduling analysis for each workstation and process performed therein to evaluate performance of the factory at production levels meeting the predetermined shipping rate; and

determining factory capacity analysis for each workstation and process performed therein to evaluate performance of the factory at production capacity levels and identify inefficient workstations that limit production line output.

17. The factory production scheduling and organization process as in claim 16 wherein the step of determining material scheduling analysis comprises the steps of:

determining the batch size of materials needed for each process performed by each workstation to meet the generated material flow rates;

determining the time required to cycle between processes for each workstation;

analyzing the required batch sizes of materials for processes and cycle time for workstations to determine optimal supply characteristics for workstations on the production line to meet the predetermined shipping rate.

18. The factory production scheduling and organization process as in claim 16 wherein the step of determining factory capacity analysis comprises the steps of:

analyzing the processing capacity of each workstation to identify the workstation operating at the highest processing capacity;

revising the material flow rates for each process in each workstation according to the highest processing capacity so as to scale material flow rates through each workstation in the production line to a capacity level;

generating, from the revised material flow rates and effective processing times for each process performed by each workstation, a revised processing capacity for each workstation reflecting processing capacity when the production line is operating at the capacity level;

determining the batch size of materials needed for each process performed by each workstation to meet the revised material flow rates;

determining the time required to cycle between processes for each workstation; and

determining the required batch sizes of materials for processes and cycle times for workstations operating at capacity level to identify inefficient workstations and processes and improvement in factory operation.

19. A method for analyzing the scheduling and organization of production for a factory, the method outputting information on optimizing factory production scheduling and organization to meet a predetermined shipping rate, said factory comprised of a plurality of workstations each performing at least one process, comprising the steps of:

receiving factory organization and operation data comprising workstation processing times and workstation set-up times for each process;

modelling the factory according to the input organization and operation data by:

(1) determining the effective processing time for each workstation to process one unit of material according to a given process;

(2) determining the effective set-up time required to configure each workstation to change between each process performed by the workstation;

(3) determining the flow rates of material in and out of each workstation for each process required to output material from the factory meeting the predetermined shipping rate; and

(4) determining the production capacity of each workstation to handle material flow rates in and out of each workstation for production of material units at the rate necessary to meet the predetermined shipping rate; and

analyzing the production performance and efficiency of the factory by:

(1) evaluating the material flow rates through each workstation for each process cycle time between processes for each workstation to identify optimal material supply requirements for each workstation necessary to meet the predetermined shipping rate;

(2) simulating factory operation at peak capacity levels by scaling material flow rates through each workstation to capacity levels and evaluating the material flow rates and process cycle times at peak capacity levels to identify inefficiently operating workstations and processes to establish improvements in factory organization and operation; and

operating the plurality of workstations to implement the established improvements in factory organization and operations on factory production facilities.

20. The production scheduling and organization method as in claim 19 wherein:

the step of determining the effective processing time comprises the step of analyzing the time required by each workstation for each process to meet production of a set number of material units; and

the step of determining the effective set-up time comprises the step of analyzing the time required by each workstation to configure for production of material units.

21. The production scheduling and organization method as in claim 19 wherein the step of calculating production capacity for each workstation comprises the step of analyzing the effective processing times and output flow rates for each workstation and process to evaluate the production capacity of each workstation to process material to meet the calculated flow rate.

22. The production scheduling and organization method as in claim 21 wherein the step of evaluating material flow rates comprises the steps of:

determining the batch size of materials needed for supplying each workstation according to each process performed to meet the calculated flow rates;

determining the cycle time required between processes for each workstation;

analyzing the required batch sizes of materials for processes and cycle times for workstations to schedule the optimal supply of materials to each workstation on the production lien necessary to meet the predetermined shipping rate.

23. The production scheduling the organization method as in claim 21 wherein the step of simulating factory operation at peak capacity levels comprises the steps of:

analyzing the production capacity of each workstation to identify the workstation operating at the highest processing capacity;

revising the material flow rates for each process in each workstation by scaling all workstation material flow rates in accordance with the identified highest production capacity;

adjusting the production capacities for each workstation according to the revised material flow rates to simulate workstation and factory operation at a peak capacity level;

determining batch sizes of materials needed for supplying each workstation to meet the revised flow rates for production at peak capacity;

determining the time required to cycle between processes for each workstation; and

analyzing the required batch sizes of materials and cycle times for production at peak capacity to identify inefficient workstations and processes and improvements in factory operation.

24. A method for improving the performance of a factory having a plurality of workstations performing one or more processes, each process operating on a batch of material, the method comprising the steps of:

determining the batch size of material required by each workstation j to perform each process i;

determining a cycle time (CT.sub.j) for each workstation j to process the material;

evaluating the cycle times and batch sizes to identify inefficient workstations and improve the processes performed by each workstation in order to reduce the batch size and cycle time of that workstation; and

operating the processes of the identified inefficient workstations to implement the improvements identified as a result of the evaluation of the cycle times and batch sizes.

Description Submit all comments and votes

FIELD OF THE INVENTION

This invention relates generally to modeling systems for manufacturing processes, and more particularly to a method and apparatus for analyzing manufacturing processes in order to schedule production and improve the manufacturing performance of a factory.

BACKGROUND OF THE INVENTION

In the global economy, domestic manufacturing enterprises are facing formidable competition from foreign companies who offer high quality goods at low prices. Domestic enterprises cannot remain competitive with foreign or even domestic companies by manufacturing goods in accordance with conventional practices. The intense present day competition necessitates rapid and, indeed, continual improvement of the methods and facilities for manufacturing goods.

There is a bewildering array of technological innovations available to enable a company to improve its manufacturing processes. An enterprise may choose to automate its factory with robots, MPR and/or computer integrated manufacturing. It may often reduce direct labor costs and improve employee performance through motivational techniques. Improvements involving automation, however, are likely to be very expensive and to demonstrably affect the continued viability of the enterprise in a competitive market. Thereafter, accurate analysis of manufacturing processes and investment in improving existing processes is critical to performance. Difficult questions have to be asked and answered about what is to be improved, how to improve it, and in what order to make the improvements for the optimal cost-efficient performance.

Unfortunately, decision makers often rely more on intuition than on an accurate analysis. Reliance on intuition, more often than not, proves to be misleading. Intuition is often misguided by outdated beliefs or misunderstandings of the principles of manufacturing. Very often intuition does not lead to lower costs or higher quality. Indeed, these older manufacturing Principles may have the reverse effect. To cite one example, most American manufacturers use a batch production method in which batch sizes for a manufacturing process are increased so as to reduce the direct manufacturing cost per unit. Contrary to intuition, however, running larger batch sizes can actually increase indirect manufacturing costs and conceal waste functions that are likely to impede efforts to lower manufacturing costs and improve the product quality. Waste functions may include, for example, excessively long set-up times for each manufacturing process, the amount of scrap produced by a process, the amount of rework that is done, the effect of machine and human down time. These waste functions necessitate, for a given volume of material, more labor, more inventory, more capital equipment, more time, and more physical space. Thus, overhead, plant and capital costs are increased with batch manufacturing. Furthermore, increased batch sizes inevitably affect quality negatively. Finally, running large batch sizes makes it more costly to build custom products that many markets demand, decreases responsiveness to changing market conditions, and slows the introduction of new products.

The executive tasked with improving the manufacturing processes of an enterprise must therefore ignore intuition and seek guidance for improving the manufacturing process with sound manufacturing principles.

One approach that overcomes misguided intuition is the "Just In Time" or "Toyota" method. The basic tenant of "Just In Time" is that an existing shipment or factory output schedule should be met with ever smaller batch sizes of the raw materials and intermediate products that make up the final product. With "Just In Time", batch sizes are made increasingly smaller until a particular workstation fails. Appropriate adjustments are then made in the manufacturing process. The "Just In Time" method replaces misguided intuition by basing improvements on reduction of batch sizes.

Although superior to intuition, "Just In Time" has its drawbacks. Improvements to a factory using the "Just In Time" method are made slowly and can result in temporary but sometimes lengthy halts in production. Since batch size reductions are necessary to gather information on which processes in the factory require the most improvement, the method is only suitable for the manufacture of products in large lots, such as automobiles. Many enterprises lack the time, money or volume of production to make them suitable candidates for improvement by the "Just In Time" method.

A tool with which to analyze a manufacturing process before making the changes to the processes is therefore required for those not able to use "Just In Time". Indeed, even those using "Just In Time" will benefit from this sort of tool.

One type of prior art tool is one that dynamically models the real-time operation of a factory. Modeling languages such as SLAM and GPSS have been successfully used to model manufacturing processes. Successful use of these languages, however, requires expert computer programming skills. Normally, those tasked with improving the manufacturing process are manufacturing executives and engineers, not expert programmers, and lack the skills necessary to apply the modeling language techniques to their particular processes.

Moreover, dynamic simulation tools suffer from other significant shortcomings. The accuracy of the program's simulation is limited by the modeler's insight and understanding of the manufacturing process. These real-time models merely simulate the movement of material through the various manufacturing processes by monitoring the size of the queues of material at various points in the factory. Apart from showing that a process in the factory has either too much material or not enough material to process, the size of the queues of material waiting to be processed do not provide information useful in determining what component of the manufacturing process should be improved or how to improve it. Multiple hypothetical runs must be made to see what effect a given set of parameter changes will have on performance. Information about what changes in the process will yield the most significant improvement therefore must be discovered by trial and error. With a very large factory, in which multiple processes are running simultaneously, the use of such programs to simulate real-time production is so difficult that it is almost impossible to predict the effects of changes in a manufacturing processes.

To improve the efficiency of manufacturing processes in a factory, an analytical tool simple and easy enough to be used by non-expert programmers is needed for accurately modelling the manufacturing process, identifying the steps or processes which are candidates for improvement, prioritizing the candidates for improvement, and determining the character and quantity of improvement.

SUMMARY OF THE INVENTION

This invention is a tool to be used for planning improvements to a manufacturing facility. The apparatus of the preferred embodiment is a specially programmed digital computer. The method of the invention is a series of steps to be implemented, in the preferred embodiment, with a digital computer.

Unlike most other prior art models of manufacturing processes, this invention does not dynamically simulate the running of the manufacturing processes. Instead, it breaks down the factory into flows of material going into and out of each workstation in the factory for each process taking place within a workstation. Each process has associated with it an effective set-up time and an effective processing time per unit of material. The total time that it takes a workstation to set-up each process and manufacture a predetermined batch of materials for every process at that workstation is called workstation cycle time. With the batch size for each process and cycle time for each workstation, as well as other values determined by the invention, production may be scheduled and necessary improvements to the factory determined.

Minimum flow rates required to meet given outputs of the factory are first determined by the invention. With the effective set-up times for each process and processing times per unit of material, batch sizes for each process and the workstation cycle times necessary to meet the required flow rates can be determined. The invention also determines the effective processing times and the effective set-up times. The effective set-up times and the effective processing times are affected by scrap and rework generated by the process and start-up scrap and start-up rework and other factors which cause the workstation to be down. The invention is capable of modelling a factory having at least one and as many as hundreds of workstations and directly determines batch sizes and workstation cycle times, as well as other values of variables indicative of the performance of the manufacturing processes within the factory. These values then serve to guide the decision maker on improving the manufacturing processes for lower cost and higher quality manufacturing.

The invention requires information concerning all of the workstations in the factory, the processes within each workstation, the inputs and outputs of each Process, the shipping rates of the manufactured units, set-up time for each process and processing time per unit of material for each process. The invention may also be provided with the following values, although they are not necessary for the model to run: the percentage of scrap generated by each process; the percentage of rework generated by each process; the rework set-up time; the rework processing time per part or unit of measure; start-up scrap of each process; start-up rework of each process; the mean time between failures of each process; the mean time between repairs of each process; the mean time between breaks for human workers for each process; the mean time on break for human workers for each process; the transport and queue time to the next process; the dollar value of the parts or materials generated by each process; the percent of rework done elsewhere for the parts or units generated by each process; and the batch size of parts or units built for each process.

The invention analyzes the manufacturing processes in two ways. First, the schedule analysis determines from all the data provided to it the minimum allowable batch sizes for each process, and other values relating to workstation performance, to be used to meet the shipping rates provided by the user. Second, capacity analysis determines from the data provided to it the minimum allowable batch sizes and the other related values based on the factory running at peak capacity and building products in the same ratios that were entered as shipping rates. The capacity analysis automatically adjusts the material flows in the factory to run at peak capacity and shows the user which workstation(s) are limiting the capacity of the factory. At the conclusion of each analysis, the user is provided by the invention with values which can be used to prioritize the workstations and the processes in need of the most improvement, to determine how much they need to be improved, and to determine what types of improvements will result in lower manufacturing costs and higher quality products. This data includes batch sizes of materials for each process, the cycle times for each workstation, percentages of time spent setting up, processing, down, and idle for each workstation the amount of trapped inventory at each workstation, the value of the scrap at each workstation, the value of each batch at each workstation, the value of the trapped inventory at each workstation, and the manufacturing cycle time through the worst case path for every product shipped.

Thus, the invention serves as a tool to be used for improving the performance of a manufacturing facility. Further, it is easily applicable to any factory having from one to hundreds of workstations by any user with a minimal amount of training. Because of its elegantly simple solution to a very complex problem, the invention is capable of being practiced even with a personal computer.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of a digital computer.

FIGS. 2a, 2b and 2c constitute a flow diagram of the steps of the present invention as carried out on a digital computer.

FIG. 3 is a flow diagram of one of the subroutines in FIG. 2 for determining the effective processing time per part for each process and the effective set-up time for each process.

FIG. 4 is a flow diagram of a subroutine in the invention that determines the percentages of time the workstation is setting up, idle, down, and processing for workstations having capacity variables greater than or equal to one (when the workstation is overcapacity).

FIG. 5 is a flow diagram for a subroutine in the capacity analysis branch of the invention that determines which workstation(s) have the largest capacity variable and sets the batch sizes if no set up time is present.

FIG. 6 is a flow diagram of another subroutine in the capacity analysis for redetermining material flow rates and capacity variables.

FIGS. 7a and 7b constitute a flow diagram for the subroutine that determines the minimum batch size required for each process, the workstation cycle time for each workstation, and the values for other variables demonstrating workstation performance.

FIG. 8 is a flow diagram for the subroutine that determines for each process at each workstation the amount of trapped inventory, the value of the batches of material, the value of the trapped inventory trapped at each workstation, and the value of the scrap generated by each process.

FIG. 9 is a flow diagram for the subroutine that determines the manufacturing cycle time of the factory.

FIG. 10 shows an example of a model factory diagram for schedule analysis.

FIGS. 11 through 13 show examples of data provided by the user to the invention of the model factory.

FIG. 14 is an example of a bill of material for a process within a workstation of the model factory.

FIG. 15 is an example of the shipping rates entered by the user for each product shipped by the model factory.

FIG. 16 is an example showing the batch sizes, batch values and scrap values determined by the invention for each process in the model factory.

FIG. 17 is a graphical representation of the sums of the batch sizes for each station in the model factory determined by the invention.

FIG. 18 is an example of the cycle times and percentage breakdown of the time spent at each station in the model factory as determined by the invention.

FIG. 19 is an example of the amount of each part trapped at each station and the corresponding part values as determined by the invention.

FIG. 20 is an example of the manufacturing cycle times as determined by the invention for each part shipped.

FIG. 21 is the worst-case path through the factory that is used by the invention to determine the manufacturing cycle time for one of the parts that is shipped.

FIG. 22 shows an example of a model factory diagram for an improved factory.

FIGS. 23 through 25 show examples of data provided by the user to the invention of the improved model factory.

FIG. 26 is an example of a bill of material for a process within a workstation of the improved model factory.

FIG. 27 is an example of the shipping rates entered by the user for each product shipped by the improved model factory.

FIG. 28 is an example showing the batch sizes, batch values and scrap values determined by the invention for each process in the improved model factory.

FIG. 29 is a graphical representation of the sums of the batch sizes for each station in the improved model factory determined by the invention.

FIG. 30 is an example of the cycle times and percentage breakdown of the time spent at each station in the improved model factory as determined by the invention.

FIG. 31 is an example of the amount of each part trapped at each station and the corresponding part value in the improved model factory as determined by the invention.

FIG. 32 is an example of the manufacturing cycle times as determined by the invention for each part shipped in the improved model factory.

FIG. 33 is the worst-case path through the improved model factory that is used by the invention to determine the manufacturing cycle time for one of the parts that is shipped.

DETAILED DESCRIPTION OF THE DRAWINGS

In the preferred embodiment, the apparatus of this invention is a specially programmed data processing system. The steps of the method of this invention are performable by a data processing system.

Referring to FIG. 1, the hardware of the preferred embodiment is a data processing system having a microprocessor 102 and a random access memory device (RAM) 104 for storing data and software.

A user interface device 106 is connected to the microprocessor 102 and memory device 104 through which the user may enter and receive data. In the preferred embodiment, the user interface device 106 is a video display monitor with an alphanumeric keyboard. The interface device may also include a printer. A data storage device 108 for storing software, factory models, and other information for implementing the invention is likewise connected to microprocessor 102 and memory 104. In preferred embodiment permanent data storage device 108 is a hard disk drive or other comparable device such as those employing magnetic or optical media. A floppy disk drive should also be available to initially install the software.

Data processing system 110 can be any digital processor having adequate processing speed and memory. A personal computer having a microprocessor operating at 10 to 33 MHz and a storage capacity of 40 Megabytes are exemplary. Many commercially available "Laptop" computers are likewise suitable.

A description of the theoretical model and of its nomenclature which forms the basis for the present invention will be helpful to an understanding of the invention. A factory has one or more workstations, each of which carries out one or more processes. The factory also includes receiving, shipping and possibly quality control stations. Every process has at least one input and output. Material to be processed originates at a receiving station. An input for a process arrives from either a receiving station, another workstation or a quality control station. The outputs of each process may be routed to another workstation, a quality control station or shipping station. The final factory outputs are all sent to the shipping station(s), where they are shipped at known shipping rates for the schedule analysis and at calculated shipping rates for the capacity analysis.

Each process at a workstation processes a particular type of material. The average time that it takes for the process to process one unit of material/is defined as the processing time per unit of material (P). The average processing time is generally taken while the worker is present, the machine is running and when rework is not required. A unit can be a discrete part or a unit of measure of the material. When a workstation has more than one process, the end of one process may require the next process to be set up. The set up time (SU) for the second process is the time between the last good unit produced by the first process and the first good unit produced by the second process less the processing time for one unit of the second part. Set up time is generally taken to be the average time to set up the process when the workers are present and the equipment is operational. Set-up produces only bad units or defective material if material is produced during set-up. Set-up material may be scrapped, the amount of scrap being referred to as set-up scrap (SUSCRP). It may also be reworked into good material, the number of units reworked being referred to as set-up rework (SUREWK). Each process may also produce a certain amount of defective material after set-up and during processing. Some of this material may be reworked and made into good material. The percentage of material to be reworked from each process is known as percent rework (%REWK). The rework may be performed at that workstation or elsewhere. The percent of parts reworked elsewhere is known as percent rework elsewhere (%REWKEL). Performing rework may require additional set-up time, known as rework set-up time (RSU). Rework can be also done at a different processing rate. The rework processing time (RP) equals the average processing time to rework a unit of material while the worker is present and the machine is operational. Those parts which were too defective to be reworked are scrapped. The percentage of parts for a given Process that are scrapped is known as percent scrap (%SCRP).

Workstations usually do not operate continually. When the workstation has material to process but it is not operating, it is due to either machine failure or human break time. The average time between machine failures is called mean time between failures (MTBF). The average time it takes to repair a workstation experiencing a machine failure is the mean time to repair (MTTR). Similarly, when a workstation whose operation depends on a human being cannot process units of material because of the absence of the human being, there is human down time. The human's absence is usually due to a break taken at specified intervals for a certain amount of time or for other functions that the operator must do that are not directly related to manufacturing. The average time between breaks is called mean time to break (MTTB), and the average time on break is called mean time on break (MTOB).

Once material has been processed by a workstation and is ready to be sent to a subsequent station, it must be transported. The material often, however, must wait to be transported to the next process. This is known as queue time. Transport and queue time are collected into a single variable for each process called transport and queue time (TPQT).

A list of the above variables and their definitions is provided in Appendix I.

With this model of the factory, the invention quickly provides to the user values of a number of variables with which the user may improve the performance of the factory. FIG. 2, a flow diagram of the subroutines comprising the application software of the present invention, shows the general method with which the present invention analyzes the above factory model. Details of the method are shown in the flow diagrams of selected subroutines shown in FIGS. 3 to 9.

Referring to FIG. 2, the invention begins with block 202. The data processing system 110 is provided by the user with factory data through user interface 106. The data is stored in memory device 104 and permanent storage device 108. Factory data includes information on all workstations, receiving stations, quality control stations, shipping stations, and all processes for each workstation and quality control station. Inputs and outputs for each process must also be provided. This information is entered by the user via a keyboard on an interactive basis. Of course other means may be used to provide the information to the digital processor such as magnetic tape, magnetic diskettes, magnetic hard disks, optical storage devices, or any other digital storage media. Indeed, once a factory model is created from this data, it may be stored in and later retrieved from the permanent storage device.

Data which must be provided to the digital processor are the processing time per unit of material and the set-up time for each process. Values for the following variables may, but do not necessarily have to be entered:

Percent scrap (%SCRP)

Percent rework (%REWK)

Rework set-up time (RSU)

Rework processing time (RP)

Start-up scrap (SUSCRP)

Start-up rework (SUREWK)

Mean time between failures (MTBF)

Mean time to repair (MTTR)

Mean time to break (MTTB)

Mean time on break (MTOB)

Transport and queue time to next process (TPQT)

Dollar value of material (VALUE)

Percent rework elsewhere (%REWKEL)

Manual batch size (MANBAT)

Two additional variables not previously defined are the dollar value of material (VALUE), being the dollar value of each part or unit of material passing through a process, and the manual batch size of material (MANBAT) for a process The latter provides the opportunity for the user to enter batch sizes of the materials actually to be used for each process instead of determining them from shipping rates.

The next step, shown in block 204, is performed for every process at each workstation of the factory, and executes a subroutine to compute the effective processing time per part (Peff) and the effective set-up time (SUeff) per process for every process. The following example and diagram is helpful to understanding Peff and SUeff. For a workstation performing two processes, A and B, its work cycle may be diagramed along a time axis with "1" representing a unit of material processed and "--" representing set up time: ##STR1##

The time required of a process to build one unit of material is equal to P. Peff for process "A", on the other hand, is the time period T2 divided by number of good units of "A" built. If some of "A" that is built is defective and must be scraped or reworked, the number of good "A" falls and Peff grows longer. Similarly, if the machine breaks during the period T2 or human beings necessary to operate the workstation go on break, T2 is longer and Peff increases. Period T3 represents the time to set up to rework defective "A's". Period T4 represents the time to rework defective material produced by the processing of "A" into good material. The time to rework a unit of material is RP. Since reprocessing produces good material it is also included with the processing time. For each unit of "A" that is reworked, RP must be added to the total processing time. The effect of rework on Peff must be taken into account when determining Peff.

Until a good unit of "A" is produced, the workstation is said to be setting up. Set up includes by definition, therefore, the time to produce the defective parts that may be incidental to setting up a process. Similarly, by definition, set up time is only affected by machine down time when the machine is producing either start-up scrap or start-up rework since the machine is not running during the rest of set-up. Human downtime, however, directly affects set up time and processing time; the absence of a human being required to set up or process parts lengthens set up time or processing time. If to rework material from process "A" requires additional set up, the rework set up time (RSU) period T3 shown in the above diagram for process "A" will effect SUeff for process "A". Details of the computation of Peff and SUeff are described in connection with FIG. 3.

The next step represented by block 206 of FIG. 2, executes a subroutine in the software for determining the flow rate of material into and out of each process at each workstation that is necessary to meet the given shipping rates. ("flow" refers to the movement of both liquid and solid material.) By the principle of conservation of matter, the amount of material supplied to a process must equal the amount of material out plus scrap. The flow rate into a process (FLOWIN) at a workstation is related, therefore, to the flow rate out (FLOWOUT) by the following equation:

Beginning with a receiving station and ending with a shipping station, material passes through various workstations and forms one or more paths. With equation (1), the input flow rate for each process within each workstation necessary to generate an output flow rate to meet shipping rates at the shipping stations are found by tracing all the paths that all the material, and their components, took through the factory. FLOWIN.