Quality Digest November

Size: px

Start display at page:

Download "Quality Digest November"

Dustin Allen
5 years ago
Views:

1 Quality Digest November

By Stephen Birman, Ph.D. I t seems an easy enough problem: Control the output of a metalworking operation to maintain a CpK of 1.33.

2 By Stephen Birman, Ph.D. I t seems an easy enough problem: Control the output of a metalworking operation to maintain a CpK of Surely all you have to do is set up a control chart to monitor some critical dimensions, set up the upper and lower control limits for those dimensions, start cutting parts, measure samples, plot them on your control chart and make adjustments when it looks like your process is going out of statistical control. It should be easy Unfortunately, it isn t. Process control in metalworking hasn t changed much during the past 100 years and is still more an art than a science. Material inconsistency, variation of tool geometry, temperature oscillation on the cutting edge and the sheer physics involved when a tool comes in contact with a work piece create a process that - unlike most processing technologies - can t be accurately predicted and controlled using statistical process control. This article discusses why this is and then demonstrates how real-time adaptive tool-wear modeling can be used in discrete part manufacturing to achieve zero defects. Quality Digest November

3 The misuse of SPC in metalremoval and forming processes SPC functions by partitioning common and special causes of variation. This partitioning is based on differences between those causes statistically and requires special statistical studies. In general common causes of variation are systematic and characterized by the normal distribution of averages. Large and nonsystematic causes are defined as assignable, or special. A statistical sense requires sufficient statistical knowldedge and time consuming studies, one of which is usually absent in the real world of machining and forming. Further, in order to provide valid results. SPC assumes that a process will be random, linear, repetitive, statistically predictible and sysmmetrical relative to nominal specifications. A metalremoval and forming process is rarely any of these. A metalworking process is a combination of a continously deteriorating tool and dynamically changing variation - both random and nonrandom. The unideirectional nature of tool wear is the main reason why generic SPC rules will never apply to close-tolerance metalworking processes; as the tool wears, the dimension moves in one direction. Generic SPC rules would flag this as a problem when, in fact, it s simply the unavoidable nature of the process. Metalworking is also nonlinear, which substantially reduces the accuracy of traditional trend analysis. Typically, a tool-wear curve consists of small, steep slopes or irregular duration within a single tool lifespan: Cutting-edge round-off and initial flank wear are low-rate processes while, in the latter stages of tool wear, flank wear and cratering might be a high-rate process. Instability of sample variation Top 12 Reasons why ipact is Gaining Ground over SPC Completeness of real-time quality documentation & assurance of data integrity Savings due to precise set-up acceptance and measurement data control Benefits of model-based feedback to an operator Gains due to computerized data interpretation of quality data Adaptability to process details & quality requirements Ability to control efficiency & quality of processes with tool wear Ability to control close-tolerance characteristics Ability to control related, tool-bonded characteristics Accuracy of capability (Cpk) control Means for precise estimation of deviation from specification limits Use of variable sampling intervals for defect prevention Ability to control a large number of quality characteristics Score is based on a survey of 16 Micronite users is affected by tool geometry and wear, and the amount is unique for every tool. SPC is based on historical data, which is almost impossible to assemble into statistically valid models for close-tolerance processes and short runs. On long runs, simply changing or sharpenning a tool would require creating new control limits. Histograms are of little help for close-tolerance applications because of their blindness to the dynamic element of process variation. A single tool run can consist of several processes, each of which would be represented by its own histogram. Multimodal representation is often misleading due to overlapping ipact SPC approximating functions. Unless SPC is being used on a process with an open tolerance and very low tool-wear rate, it simply won t work for most metalworking operations. Worse, if used on the wrong process, it will lead to erroneous results such as failing good parts and passing bad ones. The fundamental misuse of SPC in metalworking processes stems from a little change in semantics: SPC seeks a state of statistical control. People on the shop floor translate this incorrectly into a state of process control. A state of process control in discrete metal part manufacturing should be defined as a state in which the behavior of a dynamic process is Quality Digest November

predictable and product variation is balanced with equipment efficiency. If a machine shop s customers understand this difference, they will no longer demand X-bar and R charts.

4 predictable and product variation is balanced with equipment efficiency. If a machine shop s customers understand this difference, they will no longer demand X-bar and R charts. A simple metaphor illuminates the difference between SPC and process-adaptive control: Show a statistician an electrocardiogram, and he or she could likely tell you whether - given the data and the normal levels - the heart in question was performing within statistical control. If it were not, the same person could recommend that the patient take some kind of corrective action to eliminate special causes. Show the same EKG to a physician and he or she could - given a holistic examination of the patient as well as the data - recommend a specific corrective action for the particular patient s condition. If SPC is a statistician, process adaptive control is a physician. With all this in mind, the greatest challenges one who s designing a metalworking process control system face are how to recognize random and nonrandom components of variation, how to separate the mini-processes that turn up on cutting of forming tool edges and how to predict tool breakdown. Intelligent process-adaptive control technology Tool wear in an ideal metalcutting process progresses in three phases: round-off of cutting edge, low-rate tool wear and transition to high-rate tool wear. The duration of each phase isn t repeatable even in the most stable process conditions. Multiple sources of variation on the cutting or forming edge change the tool wear curve every time an insert is indexed or a form tool is sharpened. Additionally, most metalworking processes are designed to use a tool for Intelligent Control Technology Process Adaptive Control Intelligent Control Process Control Technology creating more than one dimension. It s difficult to avoid a displacement between the locations of each dimension relative to specification limits. The difference in cutting speed for associated dimensions leads to different tool-wear rates on the same tool. Roundness, taper and surface finish are all related to a tool. The objective of any process control technology is to determine the moment when a tool should be compensated for, or changed, in order to provide compliance with quailty requirements, including all Provides Reduction of quality costs Sustained improvement of process efficiency and quality Short cycle of Six Sigma implementation Achievement of operators consistency Allows Use of dimensional data for predictive process control Use of modeling for improvement of efficiency and quality Prevention of tool breakdown and excessive wear Fine-tuning of network for particular operations Conducts Diagnosis of equipment and measuring system Real-time data analysis with decisions on process action Assessment of product quality for non-random processes Estimation of capability to hold a tolerance Offers Company-wide total data control Process and quality database for machine maintenance Estimation of process and tool design Cross-functional problem management dimensions and tolerances created by this tool. Currently, predictive modeling is the only way to accurately control such processes. MICRONITE, an expert system based on the ipact concept and developed by High Tech Research Inc., of Deerfield Illinois, uses predictive modeling as part of its intelligent process adaptive control technology. There are four major components of variation. The following causes of variation identified for each component can be found in any metalwork- Unidirectional tool wear and random and non-random variation makes any discrete process as Non-linear with small and steep slopes of irregular duration within a tool lifespan Non-repetitive due to inconsistencies in machinability, tool geometry, etc. Asymmetrical relatively to nominal due to intended and non intended shifts in process centering Unstable in sample variation within and between tool runs Only the process adaptive model can accurately control unique combinations of these components Quality Digest November

ing process: Basic sample variation Includes measurement error (e.g., accuracy, precision and repeatability), shape variation (e.g., taper and roundness) and machine precision (e.g., piece-to-piece variation, inter-spindle and interfixture variation).

5 ing process: Basic sample variation Includes measurement error (e.g., accuracy, precision and repeatability), shape variation (e.g., taper and roundness) and machine precision (e.g., piece-to-piece variation, inter-spindle and interfixture variation). The extent of this variation determines whether the tolerance is wide-open, open, close or extremely close. As an example, a close tolerance on a multispindle screw machine becomes an open tolerance on a CNC lathe. Quantifying primary variation helps uncover causes and reduce variations that force frequent and unnecessary adjustments and tool changes. Process-dependent variation Reflects tool wear and the instability of a dynamic system. These include nonrandom variation due to tool-wear trend and random variation due to the instability of process variables. Cumulative product variation Depends on the alignment of process runs. These include variation due to the displacement of the locations of sample averages between sequential processes and variation caused by indiscriminate process interruption and adjustment. Special causes of variation as defined by SPC Variation related to machine, material and people. Knowledge of the dynamic nature of discrete processes helps us understand different ways to achieve a state of process control. Adaptability is where it all begins. An intelligent system should be capable of controlling a multitude of relatively stable and unstable processes at once. The system should cope with a range of quality requirements, from compliance-to-specifications to Six Sigma. This means that unique sampling design and automatic execution of predictive modeling is required for every operation and characteristic. MICRONITE uses three types of predictive models: nonlinear trend control, control of probability of defects and control of extent of tool wear. Because sampling time is critical, adaptive intervals for every tool are updated after each data entry. The system also allows pre - determined levels of product variation. Real -time control by a CpK model (without using X - bar and R charts) guarantees a customer -required variation index. How does it work? A workstation is located near a machine, group of machines or a cell. Upper and lower specification limits are entered for each dimension, as is the cycle time for each cutting operation. Based on the tool - wear model, the number of parts run and other parameters, the software prompts the user to measure a unit or sample. The software then tells the operator to continue, input a compensation adjustment into the machine or change the tool. It will also tell the operator how long before the next measurement must be taken. The software adapts the model as new data is collected Quality Digest November

Process adaptive models for real-time control of metalworking operations Control models providing compliance with specification Predictive tool wear control Control of risk of defects Short-run trend

6 Process adaptive models for real-time control of metalworking operations Control models providing compliance with specification Predictive tool wear control Control of risk of defects Short-run trend & variation control Predictive control of unilateral tolerances One-sided & two-sided pre-control Engineered process control models Control of extent of tool wear Control of product variation Six Sigma & centering control Cpk control for shifted processes Statistical process control (SPC) Control of extent of tool wear Expert Band Control limits the extent of tool wear in order to avoid expensive tool repair. The model allows the process to shift from the spec nominal. In this case, if process would not be stopped as the system demands, acceleration of tool wear would ruin the tool. Six Sigma and centering control MICRONITE guides an operator to comply with the most demanding requirement for process control: compliance with Six Sigma and centering around the nominal. In this example, MICRONITE kept a process around the nominal and recognized the beginning of accelerated phase of tool wear. A process was stopped before a sharp edge would become dull Here is a an oversimplified description of what actually occurs. First, a process (i.e., tool wear) curve is segmented in real -time by one of ipact's rule -based models. After every data entry (either sample or unit), a model predicts the risk of defects until the next inspection, the rate of tool wear related to the location of data averages, and sample variation. If the system determines that a risk of defects has increased, a compensation adjustment will be recommended; if tool -wear rate has dangerously increased, a tool change is demanded (see Fig 1). A trend control model will stop a process at the point of accelerating tool wear and increased risk of defects. A model controlling the extent of tool wear will stop a process before severe tool deterioration occurs. Stable processes with relatively low tool - wear rates are controlled by separate estimates of the probability of defects on the lower and upper sides of specification limits. Sampling time is critical and is adjusted dynamically after each measurement; a stable process with little variation would require a longer sampling interval than a process with large variation and high -rate tool wear. When a tool is indexed, sharpened or changed, you can t expect duplication of a process curve; all tool -wear processes are nonrepetitive. This means that all parameters, such as sample variation, tool -wear slopes, rate of tool wear acceleration and a number of mini - processes, will probably change. Therefore, a new model is needed in order to control a new process. Whenever the user indicates a process change, ipact will start a new modeling cycle. Only when mature and long - running processes repeat are historical data used for process control decisions. Predictive trending and adaptive sampling intervals ensure that a machine is operated to its maximum capacity. Tools achieve maximum usage before change -out, and sampling is performed only when needed. Both of these decrease waste and increase machine uptime and operator efficiency. Control of operations with multiple tool -bonded processes Let s look at how predictive modeling works with different tools and process types. An ideal condition for process control exists when a finishing tool creates only one critical characteristic and a process can be easily adjusted to nominal specifications. If metal cutting were only this, SPC would be a solution. However, most of metal - removal and - forming operations are designed to create multiple characteristics with a single finishing tool. These multiple tool -bonded processes are divided between the following: Type one Single -point and simple -shape tool (e.g., insert, mill or reamer). Individual characteristics are cut sequentially using the same tool. A process control solution must be able to control for all misaligned tool - bonded characteristics and control primary variation. Type two Step tool (e.g., form tool, step drill or grinding wheel). Individual characteristics are cut simultaneously. A process control solution must control for all misaligned tool -bonded characteristics, different tool - wear rates and primary variation. Type one is considered a single process with multiple outputs (i.e., one tool cutting several dimensions). Type two is considered multiple processes with multiple outputs (i.e., a step - wise tool generating related dimensions and tolerances). Tool - bonded processes of type one can be controlled by a key characteristic if misalignment and primary variation aren t significant. It s here that SPC can be used effectively. Typically, however, a close - tolerance metalworking operation faces the problem of Quality Digest November

Synchronized control of related tool-bonded characteristics Maximum process efficiency and zero-defect quality are guaranteed by MICRONITE decisions to re- misalignment between related

7 Synchronized control of related tool-bonded characteristics Maximum process efficiency and zero-defect quality are guaranteed by MICRONITE decisions to re- misalignment between related characteristics and a substantial primary variation. SPC can't handle this. In this case, ipact uses multiple models to control each critical product characteristic, one model per characteristic. Step tools, or type two, are even more problematic, having two or more critical characteristics with different dynamic patterns in terms of tool -wear rates, variation and the location of sample averages. If more than one process is running on a complex tool -cutting edge, then ipact uses a multiprocess model. This forces the optimization of internal specifications for tool -bonded characteristics. It also helps develop solutions for maximum tool efficiency. Six -level data organization Although we ve been considering the use of predictive mod- sharpen a shave tool which creates four OD s. eling at the part level, it s most effective when applied holistically. This is achieved when all workstations are networked and data is shared upstream. The basic pattern of data organization is divided into six levels. Production plant At this level, the software provides an overview of all jobs and operations and alerts for problems. Multioperation job Here, data continuity is provided for all operations. Metal-cutting or-forming machines The equipment's capability to hold tolerance is verified by in -process capability control and off -line studies. Operation Control of individual and tool-bonded characteristics along with roughing and finishing tools is aimed at increased efficiency and limited machine attendance. A group of tool-bonded characteristics Reducing misalignment and variation leads to an extended time between tool adjustments and changes. Individual characteristics Compliance with specification, a sampling discipline and preventing tool failure must be achieved. We ve only briefly discussed this exciting new technology, which is currently the only viable solution for in-process metalcutting and-forming control; it does what SPC cannot. Considering that in 1998, metal-cutting and-removal operations accounted for between $240 billion and $850 billion of domestic expenditure, billions of dollars per year could be saved through waste reduction and increased machine and operator efficiencies using this technology.. Quality Digest November

The Statistical Cracks in the Foundation of the Popular Gauge R&R Approach

The Statistical Cracks in the Foundation of the Popular Gauge R&R Approach 10 parts, 3 repeats and 3 operators to calculate the measurement error as a % of the tolerance Repeatability: size matters The