Process control system and method

• US 20030033040A1
• Filed: 05/29/2002
• Published: 02/13/2003
• Est. Priority Date: 05/29/2001
• Status: Active Grant

• Current Assignee: John Billings
• Original Assignee: John Billings
• Inventors: Billings, John
• Granted Patent: US 7,149,597 B2
• Time in Patent Office: Days
• US Class Current: 700/97
• CPC Class Codes: G05B 13/021 in which a variable is automatically adjusted to optimise the performance
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202

Abstract

A process control method for producing acceptable product within a range of acceptable specifications and unacceptable product that are not within the range of acceptable specifications to optimize profit and utilization of resources, the process having at least one input and at least one output that is affected by variation in at least one input, the method including varying the input about a mean of at least one acceptable specification so that the output is optimized to where the revenue from producing one more acceptable unit of product is substantially equal to the cost of producing one more unacceptable unit of product. In another embodiment, a method is provided that includes optimizing a process having inputs and outputs by measuring the inputs and outputs; determining variation of the mean (u_of) and variation about the mean (u_abt); determining causes of u_of; and controlling u_of to match u_abt data values. The method would further include determining the relationship between the output defect rate and changes in inputs, control of u_of; modeling the relationship; and solving for optimum set point of one or more of the inputs. Controlling u_of can include adjusting the inputs, and solving for optimum set point ideally involves utilizing functions for input defect rate, revenue, and cost

Claims

1. A method of controlling a process for producing acceptable product within a range of acceptable specifications and unacceptable product that are not within the range of acceptable specifications, the process having at least one input and at least one output that is affected by variation in at least one input, the method comprising:

varying the at least one input about a mean of at least one acceptable specification so that the at least one output affected by the at least one input is optimized to where the revenue of producing one more acceptable product is substantially equal to the cost of producing one more unacceptable product.

2. A method of optimizing a process having at least one output that varies in response to variation of at least one input, the method comprising:

determining variables of at least one input and at least one output to measure;

measuring the variables of the at least one input and the at least one output while the variables are adjusted in the course of performing the process;

determining variation of the mean and variation about the mean for each of the measured variables of the at least one input and the at least one output;

determining an adjustment in the variables in the at least one input required to bring the mean of the variables in the at least one output within a target value range;

determining the cost of making one more unacceptable product that is equal to the revenue from making one more acceptable product; and

adjusting the variables in the at least one input to achieve an optimum setpoint of at least one selected variable of the at least one input.

3. A method of optimizing a manufacturing process, comprising:

providing a plurality of components configured to be assembled together into a single unit; and

matching components that are under tolerance to components that are over tolerance to achieve a single unit that is within tolerance.

5. The process of claim 3, wherein the first subcomponent is under the mean tolerance to the same degree that the second subcomponent is over the mean tolerance.

4. A manufacturing process, comprising:

matching a first subcomponent that is under a mean tolerance with a second subcomponent that is over the mean tolerance to achieve a component formed of the first and second subcomponents that is at the mean tolerance.

6. A method for controlling a process having one or more inputs and one or more outputs where outputs represent characteristics and definition of the product, and inputs are materials and process operations that produce outputs, the method comprising:

identifying at least one output and corresponding inputs that when varied affect the output;

measuring the identified at least one output and corresponding inputs;

determining a variation of a mean and a variation about the mean for each of the at least one output and inputs; and

determining an optimum value of the corresponding inputs to control the at least one input to a target point that represents where the marginal cost of producing one more defective product equals the marginal profit of producing one more non-defective product.

7. A method for controlling a process to produce acceptable product, comprising:

adjusting a process input about a mean of measurements of the process input to the point where the marginal cost of producing one more unit of defective product is equal to the revenue from producing one more unit of non-defective product.

8. A method for controlling the production of engineered wood product to maximize profits, comprising:

assembling engineered wood;

bonding the engineered wood with a bonding agent in a press that applies heat and pressure to the engineered wood product and the bonding agent for a predetermined cycle time;

measuring the bonding of the engineered wood product using a bond measurement system that generates a bond measurement signal; and

adjusting at least one of the pressure, temperature, and cycle time of the press in response to at least the bond measurement signal to permit defective products to be produced to the point that profits are maximized.

9. The method of claim 8, further comprising:

marking acceptable product with indicia indicating the product is not defective and sorting the defective products from the acceptable products.

10. A method for controlling the production of engineered wood product to maximize profits, comprising:

assembling engineered wood;

bonding the engineered wood with a bonding agent in a press that applies heat and pressure to the engineered wood product and the bonding agent for a predetermined cycle time;

measuring the thickness of the engineered wood product using at least one thickness measurement device that generates a measured thickness signal;

measuring the bonding of the engineered wood product using a bond measurement system that generates a bond measurement signal;

adjusting at least one of the pressure, temperature, and cycle time of the press in response to at least the bond measurement signal to permit defective products to be produced to the point that profits are maximized;

marking acceptable product with a mark indicating the product is not defective; and

sorting defective product from acceptable product.

12. The system of claim 10, further comprising a marking device configured to mark acceptable product with indicia to indicate the product is not defective and a device to sort acceptable product from defective product.

11. A system for producing engineered wood, comprising:

a device for forming sheets of wood;

a press for bonding the sheets of wood with a bonding agent with the application of heat and pressure for a cycle time to form engineered wood product;

a bond integrity inspection system configured to measure the bond strength of the product and to generate a bond measurement signal; and

a process controller configured to receive the bond measurement signal and to adjust at least one of the heat, pressure, and cycle time of the press to maximize profits while permitting defective product to be produced.

13. A method of optimizing a process having inputs and outputs, comprising:

measuring the inputs and outputs and saving data therefrom;

determining u_of and u_abt from the data;

determining causes of u_of; and

controlling u_of to shift u_of to match u_abt data values.

14. The method of claim 13, further comprising:

determining the relationship between output defect rate and selected variables of at least one input;

modeling the relationship; and

solving for optimum set point of the selected variables of the at least one input.

15. The method of claim 13, wherein controlling u_of comprises adjusting inputs.

16. The method of claim 14, wherein solving for optimum set point comprises utilizing functions for input and output variation, output defect rate, revenue, and cost.

17. A method of solving for changes in multiple input variables so that the combination of input changes results in an optimization of output, comprising:

solving for a combination of input changes that optimizes output based on defect rates, cost and selling price by first solving for changes in individual inputs and the resulting change in output based on variation, defect rate, cost, volume and selling price, and then solving the changes in the individual inputs simultaneously to arrive at a set of optimum input and output target set-points.