WO2017058156A1

WO2017058156A1 - Equipment set up time prediction model creation

Info

Publication number: WO2017058156A1
Application number: PCT/US2015/052854
Authority: WO
Inventors: Sunil KOTHARI; Thomas J PECK; Jun Zeng; Gary J DISPOTO; Michael L. REASONER; Francisco OBLEA
Original assignee: Hewlett-Packard Development Company, L.P.
Priority date: 2015-09-29
Filing date: 2015-09-29
Publication date: 2017-04-06

Abstract

Examples disclosed herein relate to creating an equipment set up time prediction model. In one implementation, a processor determines equipment set up experimental operations to indicate equipment set up changes associated with changes in variables. The processor may create an equipment set up time prediction model based on the results of the equipment set up experimental operations such that the equipment set up time prediction model is based on set up time and variable correlation information. The processor may output information related to the equipment set up time prediction model.

Description

EQUIPMENT SET UP TIME PREDICTION MODEL CREATION

BACKGROUND

[0001] A manufacturing environment, such as a print shop or additive manufacturing shop, may involve multiple pieces of equipment to produce multiple types of products and product parts. For example, a customized printed item may be produced in multiple stages, including print related stages and binding related stages. Likewise, an additive manufacturing environment may include additive manufacturing stages as well as stages related to post processing related to the item created using additive manufacturing techniques. Different types of equipment may be associated with the different stages.

BRIEF DESCRIPTION OF THE DRAWINGS

[0002] The drawings describe example embodiments. The following detailed description references the drawings, wherein:

[0003] Figure 1 is a block diagram illustrating one example of a computing system to create an equipment set up time prediction model.

[0004] Figure 2 is a flow chart illustrating one example of a method to create an equipment set up time prediction model.

[0005] Figures 3A-3C are diagrams illustrating examples of creating an equipment set up time prediction model.

DETAILED DESCRIPTION

[0006] A manufacturing environment, such as a print shop or additive manufacturing shop, may involve multiple products and/or product parts created successively on the same piece of equipment such that the set up time between products or sets of products may have an impact on manufacturing time as well as downstream manufacturing activities in instances where multiple pieces of equipment are used. In one implementation, a processor creates an equipment set up time prediction model based on a variable. For example, changes between product variables related to successive equipment operations may result in a change in the equipment set up. A processor executing a software product may determine equipment set up experimental operations to be performed such that the results indicate equipment set up times associated with changes in variables. The equipment experimental operations may be operations identified as operations that would provide information to be used in creating and/or updating the equipment set up time prediction model, such as set up time information associated with missing, incomplete or unverified information. The processor may create the equipment set up time prediction model based on known information, such as based on a production log. Other equipment set up time information may be estimated, and the processor may identify an experiment to update the estimation based on results of an experiment related to the set of variables. The processor may create an equipment set up time prediction model based on results of the equipment set up experimental operations such that the equipment set up time prediction model is based on set up time and variable correlation information.

[0007] The processor may output information related to the equipment set up time prediction model. An equipment set up time prediction model may be particular applicable to short run manufacturing environments for producing customized products. The creation of customized products may involve more changes in equipment features due to the increased number of types of products, and as a result, the equipment feature changes may have a larger effect on the manufacturing schedule. The equipment set up time prediction model may be used for simulations, scheduling, or other manufacturing environment analysis.

[0008] Figure 1 is a block diagram illustrating one example of a computing system to create an equipment set up time prediction model. The equipment set up time prediction model may indicate a likely equipment set up time based on product production information, such as based on a product feature. The computing system 100 includes a processor 101 , machine-readable non-transitory storage medium 102, and storage 107.

[0009] The storage 107 may be any suitable storage for communicating with the processor 101. The storage 107 may be included within the same electronic device as the processor 101 or may communicate with the processor 101 via a network. The storage 107 may include equipment set up time prediction model 108. For example, the equipment set up time prediction model 108 may receive information related to a product variable or change in a product variable as input information, and a processor may apply the equipment set up time prediction model to output information related to an estimated set up time for a piece of equipment The equipment set up time information may be used to improve the manufacturing environment, such as to create simulations, plan an equipment sequence, select a specific piece of equipment, group products for a piece of equipment, and/or select an order for a set of products to be processed by a piece of equipment. The equipment set up time prediction model 108 may take into account relationship information between variables, such as where the set up time for adjusting for two variables is different than the combined individual time of adjusting for each of the variables alone. The equipment set up time prediction model 108 may take into account how the equipment set up process occurs in the set up time, such as whether an operation occurs serially or in parallel to another operation performed on the product.

[0010] The processor 101 may be a central processing unit (CPU), a semiconductor-based microprocessor, or any other device suitable for retrieval and execution of instructions. As an alternative or in addition to fetching, decoding, and executing instructions, the processor 101 may include one or more integrated circuits (fCs) or other electronic circuits that comprise a plurality of electronic components for performing the functionality described below. The functionality described below may be performed by multiple processors.

[0011] The processor 101 may communicate with the machine-readable storage medium 102. The machine-readable storage medium 102 may be any suitable machine readable medium, such as an electronic, magnetic, optical, or other physical storage device that stores executable instructions or other data (e.g., a hard disk drive, random access memory, flash memory, etc.). The machine-readable storage medium 102 may be, for example, a computer readable non-transitory medium. The machine-readable storage medium 102 may include equipment set up time prediction model deficiency identification instructions 103, equipment set up performance instructions 104, equipment set up time and variable correlation determination instructions 105, and equipment set up time prediction model update instructions 106.

[0012] The equipment set up time prediction model deficiency identification instructions 103 may include instructions to identify a deficiency in the equipment set up time prediction model 108. The deficiency may be, for example, related to a variable and set up time correlation. The deficiency may be related to a particular set of variables and a set up time correlation, such as where variables A, B, and C are changed between product runs. The deficiency may be related to a variable set up time correlation based on estimated or incomplete information or based on an amount of data below a threshold. For example, a production log may be used to create an initial equipment set up time production model, and a particular variable to equipment set up may occur once in the log while a second variable to equipment set up occurs 100 times in the log. The equipment set up time production model may be used with some portions of the data estimated, and the processor may identify areas where an equipment set up procedure may be used to improve the accuracy of the equipment set up time prediction model.

[0013] The equipment set up performance instructions 104 may include instructions to cause an equipment set up procedure to be performed. For example, the processor 101 may identify a variable, type of equipment, and/or operation sequence associated with the equipment set up performance. The equipment set up performance instructions 104 may include instructions related to a time, location, or piece of equipment associated with the equipment set up performance. In one implementation, the equipment set up performance instructions 104 include instructions to select information about the equipment performance based on scheduled operations and/or requested products. For example, a piece of equipment that is up and running for another purpose or a time associated with a down time on the equipment may be selected for the equipment set up performance. The equipment set up performance may occur across multiple pieces of equipment and/or multiple sites. For example, a first piece of equipment may perform a first set up operation, and a second piece of equipment to perform a second set up operation, in one implementation, the processor 101 aggregates information across multiple sites. For example, the processor may create a list of equipment set up procedures and cancel procedures once they have been performed at other sites.

[0014] The equipment set up time and variable correlation determination instructions 105 may include instructions to determine the set up time and variable correlation information is based on the result of the equipment set up operation. The set up time information may be based on multiple variables. For example, the set up time and variable correlation information may be determined related to how the set up time related to the identified variable impacts an overall set up time to produce a product with multiple variables.

[0015] The equipment set up time prediction model update instructions 106 may include instructions to update the equipment set up time prediction model 108 based on determined variable and set up time correlation information. For example, a portion of the equipment set up time prediction model 108 associated with the identified deficiency may be updated based on the correlation information. In one implementation, the equipment set up time prediction modei 108 is used by the processor 101 or another process to adjust or predict production associated with a manufacturing environment. The updates to the model may occur in the background such that the updates are transparent to the application using the equipment set up time prediction model 108.

[0016] Figure 2 is a flow chart illustrating one example of a method to create an equipment set up time prediction model. The equipment set up time prediction model may be useful for predicting the time used to make a change to a piece of equipment to accommodate a change in a product variable between product operations. The equipment set up time prediction model may be used, for example, for simulations for a manufacturing environment, for scheduling production in a manufacturing environment, and/or for selecting equipment in a manufacturing environment. The method may be implemented, for example, by the processor 101 of Figure 1.

[0017] Beginning at 200, a processor determines an equipment set up time prediction model such that the prediction model accounts for equipment set up time related to a variable. The equipment may be any suitable equipment, such as equipment in a manufacturing environment. The equipment may be associated with a customized or short run production and may be associated with a print shop or additive manufacturing. The equipment set up time may be related to a single piece of equipment or a set of equipment for creating a product. In one implementation, there are multiple options for equipment to produce the same product type, and the types of equipment may be represented in the equipment set up time prediction model.

[0018] The change in the production variable represented by the equipment set up time prediction model may be any suitable variable change, such as a change related to the production environment or a change in a product variable between successive products produced on a piece of equipment. In one implementation, die processor determines whether an amount of change or percentage of change between the product variables associated with the successive products produced on a piece of equipment is above a threshold. For example, changes below the threshold may not involve a change to an equipment setting or may involve a minimal change. The processor may associate a different threshold with different variable types and/or equipment types. The changes in the variables may related to different product types and/or changes in product features associated with the same product type.

[0019] The processor may determine the equipment set up time prediction model based on any suitable information, in one implementation, the processor creates the model based on a past production log information associated with past production. The processor may create the equipment set up time prediction model based on data associated with individual pieces of equipment or based on how the pieces of equipment functioned in a different manufacturing environment, such as in a different site. The equipment set up time prediction model may take into account relationships between product variables and/or pieces of equipment. For example, there may be a serial or parallel relationship between set up time for two variables.

[0020] Figures 3A-3C are diagrams illustrating examples of creating an equipment set up time prediction model. Figure 3A illustrates one example of an equipment set up time prediction model 300 shown in a tree structure where the nodes 301-312 represent different changes to a set of production variables. The equipment set up time prediction model 300 shows relationships between changes associated with variables A, B, and C. For example, the equipment set up time may be serial or may have relationships based on the individual variables, such as where the set up time for B is shorter if a set up for variable A is also performed because a preliminary procedure related to variable B is performed in conjunction with the set up for variable A. The equipment set up time prediction model 300 may be created by a processor to represent dependencies between changes related to multiple variables. The processor may associate set up time information with the nodes 301-312.

[0021] Referring back to Figure 2 and continuing to 201, a processor identifies a deficiency in the prediction model related to a particular variable. For example, the prediction model may include some information that is based on manufacturing log and experiment data and other information that is estimated or not accounted for. For example, the prediction model may be both a black box and white box model, such that some parameters have known interaction information and others are estimated, in one implementation, the equipment set up time prediction model includes correlation information between a variable and set up time, and there is a confidence level associated with the variable and set up time relationships such that the processor identifies a deficiency based on a confidence level below a threshold.

[0022] The processor may identify the deficiency in any suitable manner, such as based on an analysis of a tree data structure indicating levels of known data associated with variables accounted for in the equipment set up time prediction model. The processor may identify a deficiency such that at least one variable in a set of variables is not associated with an equipment set up time, in one implementation, the processor identifies multiple deficiencies. The processor may prioritize the deficiencies based on the type of deficiency and/or based on an operation associated with resolving the deficiency.

[0023] Figure 3B is a diagram illustrating one example of a diagram indicating options related to set up time methods according to whether different changes in variables are known and/or have a confidence level above a threshold. For example, a processor may have a method for estimating an equipment set up time for a change in variable A and B where the information for variable A is based on previous data and the information related to variable B is estimated. For example, variable B information may be based on other similar variables or other information related to other variables. The data may be represented in any suitable manner, such as in the tree structure shown in Figure 3B. The blocks in Figure 3B represent whether variables and combinations between variables are known as well as whether relationship between variables is known, such as whether a set up time relationship is serial or parallel. A processor may analyze data related to known and unknown data to determine a deficiency in an equipment set up time prediction model, such as the model 300 in Figure 3A. A processor may analyze the model 313 to determine that an equipment experiment may be run to provide more complete information associated with node 314 related to changes in both variables A and C between two products and/or product parts. [0024] Referring back to Figure 2 and continuing to 202, a processor causes an equipment set up procedure related to the variable to be performed. The processor may cause the procedure to be performed in any suitable manner. For example, the processor may store information about procedures to be performed, such as in a scheduling log and/or may transmit information to a specific piece of equipment related to the equipment set up procedure.

[0025] The processor may determine an equipment set up procedure related to the variable in any suitable manner. The processor may determine a set of operations related to a variable, such as related to how other variables affect a set up time associated with the particular variable. The variables may be product variables, such as related to dimensions, and/or other types of parameters related to the manufacturing environment, such as related to temperature, staffing, or amount of available equipment.

[0026] The processor may determine the set up procedure based on a particular variable within a set of variables. For example, the processor may determine that the deficiency is related to a relationship between a set of variables. The processor may determine the set up time between multiple variables as a group or individually. The processor may use the set up procedure to determine the relationship between the time increases associated with the variables and whether they occur serially or in parallel. The processor may set up the procedure to determine sequential information related to the set up time, such as where the set up time is faster performing A and then B as compared to B and then A.

[0027] The processor may select location information associated with the set up procedure, such as based on workload at different sites. The processor may determine a set up operation such that different portions of the operation occur in different locations, such as different manufacturing sites, and are performed by different equipment.

[0028] The processor may select timing information associated with the set up procedure. The processor may determine a start time and/or time window associated with the set up procedure. The processor may select the time based on operation or scheduling information associated with a piece of equipment selected for the set up procedure. For example, the processor may cause the equipment set up procedure to be performed based on non-operational time associated with the equipment on which the procedure is to be performed.

[0029] In one implementation, the processor causes a user interface to be displayed to receive user input related to the operation to be performed, such as information related to a prioritization of deficiencies, a selection of an operation from a set of operations, a selection of time, and/or location information associated with a potential operation. The processor may display information about a potential operation and allow a user to accept or change information related to the potential operation.

[0030] Figure 3C is a diagram illustrating potential equipment set up operations that may be recommended by a processor to remedy a deficiency identified base on the node 314 from the model 313 in Figure 3B. For example, block 318 shows a set of three potential equipment set up experiments that may be determined by a processor related to variables A and C.

[0031] Block 315 includes information related to an identified experiment for piece of Equipment X to make adjustments to variables A and C. The experiments may include multiple types of changes related to the variable. For example, the processor may determine common dimensions or likely combinations of values.

[0032] Block 316 includes information related to equipment set up procedures performed over multiple pieces of equipment, such as where set up for a change in variable A is simulated by piece of equipment X and where set up for a change in variable C is simulated by piece of equipment Y. The equipment set up procedures may be performed over multiple sites, such as where piece of equipment X is at one location and piece of equipment Y is at another location.

[0033] Block 317 includes information related to equipment set up procedures performed where a specific time is associated with the set up procedure. For example, the set up procedure is scheduled for equipment X to make an adjustment relative to variables A and C during a scheduled downtime, in one implementation, the processor includes the set up procedure in a queue, and the processor selects the set up procedure in the queue. The processor may determine a time estimate associated with the set up procedure, and the processor may use the estimate to select a scheduled down time long enough for the estimated procedure time. The processor may determine multiple types of equipment set up procedures and/or select the set up procedures within a set of constraints, such as particular equipment or time frames.

[0034] Block 319 shows the result of an experiment set up procedure is that die set up time is adjusted by (3 x CHANGE IN VARIABLE A) + (4 x CHANGE IN VARIABLE C) such that the variable changes each independently increase die set up time.

[0035] Referring back to Figure 2 and continuing to 203, a processor updates the equipment set up time prediction model based on a result of the procedure. For example, the processor may update the equipment set up time prediction model 300 in Figure 3A block 303 with the information related to the change in variable C. The processor may update the set up time prediction model 300 block 301 to include information about the set up time for a change in variable A in conjunction with a change related to variable C.

[0038] The processor may update the equipment set up time prediction model in any suitable manner. The processor may aggregate information from multiple set up time operations to update the equipment set up time prediction model. The processor may update a confidence level associated with a portion of the equipment set up time prediction model based on the new data from the set up operation. The processor may identify factors associated with different set up times for the same variable, such as where set up time to adjust dimension A by X is affected by an additional factor, such as power level.

[0037] The processor may output information related to the equipment set up time prediction model by storing, displaying, and/or transmitting the information, in one implementation, the processor receives information about a product and/or production environment and outputs information related to an estimated set up time based on the equipment set up time prediction model. The set up time may be factored into the production time and used in simulations and scheduling related to the manufacturing environment.

Claims

1. A computing system, comprising:

a storage to store information related to an equipment set up time prediction model, wherein the equipment set up time prediction model is based on correlation information between variables and equipment set up time; and

a processor to:

identify a deficiency in the equipment set up time prediction model related to a variable;

cause a piece of equipment to perform a set up operation related to the identified variable;

determine the set up time and variable correlation information for the identified variable based the results of the performance; and

update the stored equipment set up time prediction model based on the determined set up time and variable correlation information.

2. The computing system of claim 1 , wherein the equipment set up time

prediction model takes into account set up operations that occur in parallel and set up operations that occur serially.

3. The computing system of claim 1 , wherein determining the set up time and variable correlation information comprises determining the set up time and variable correlation information related to how the set up time related to the identified variable impacts an overall set up time to produce a product related to multiple variables.

4. The computing system of claim 1 , wherein causing the piece of equipment to perform the set up operation comprises causing a first piece of equipment to perform a first set up operation and causing a second piece of equipment to perform a second set up operation, and wherein determining the set up time and variable correlation information is based on the result of the first and second set up operations.

5. The computing system of claim 1 , wherein the processor is further to: determine a piece of equipment to cause to perform the set up operation; and

determine a time frame for performing the set up operation based on a schedule associated with the piece equipment.

6. A method comprising:

determining, by a processor, an equipment set up time prediction model, wherein the prediction model accounts for equipment set up time related to a variable;

identifying a deficiency in the prediction model related to a particular variable;

causing an equipment set up procedure related to the variable to be performed; and

updating the equipment set up time prediction model based on a result of the procedure.

7. The method of claim 6, wherein the equipment set up time prediction model takes into account serial and set up time relationships between variables.

8. The method of claim 6, wherein identifying the deficiency comprises

identifying a deficiency such that at least one variable in a set of variables is not associated with an equipment set up time.

9. The method of claim 8, wherein identifying a deficiency is based on an

analysis of a tree data structure indicating levels of known data associated with variables accounted for in the equipment set up time prediction model.

10. The method of claim 6, wherein causing the equipment set up procedure to be performed comprises causing a first equipment set up procedure to be performed at a first site and a second equipment set up procedure to be performed at a second site.

11. The method of claim 6, wherein causing the equipment set up procedure to be performed comprises determining a time frame to cause the equipment set up procedure to be performed based on non-operational time associated with the equipment on which the procedure is to be performed.

12. A machine-readable non-transitory storage medium comprising instructions executable by a processor to:

determine equipment set up experimental operations to indicate equipment set up changes associated with changes in variables;

create an equipment set up time prediction model based on the results of the equipment set up experimental operations, wherein the equipment set up time prediction model is based on set up time and variable correlation information; and

output information related to the equipment set up time prediction model.

13. The machine-readable non-transitory storage medium of claim 12, wherein instructions to determine equipment set up experimental operations comprise instructions to determine an experimental operation based on a deficiency of the equipment set up time prediction model of information related to a variable within a set of variables .

14. The machine-readable non-transitory storage medium of claim 12, wherein creating the equipment set up time prediction model comprises instructions to: determine set up time dependencies between equipment set up time and multiple variables; and

create the set up time prediction model to account for the time dependencies.

15. The machine-readable non-transitory storage medium of claim 12, wherein the equipment is related to at least one of: additive manufacturing and printing.