The new generation process and production control system described in this book is used to control all activities on machine shop floor and to provide data-driven engineering solutions. Several key functions are required to control production activities in a modern plant. These functions can be grouped into six categories: (1) manufacturing planning and process control design, (2) data capture and intelligent data processing, (3) analytical process visualization and automated decision making, (4) low-risk assurance of quality , (5) optimization of process and production conditions, and (6) dynamic control of production efficiency and effectiveness. For any particular case, it is not necessary or feasible to include all functions. Rather, different manufacturing plants require different sets of functions and control tools depending on equipment and the production structure.

The main function of the shop control system is to capture sufficient amount of inspection / production data and automate decisions at the point of data entry. This function is implemented through embedded machining knowledge managed by the expert system. Operator interface rules relate to events which require actions, such as a sampling time, tool offsets, and tool changes. Quality assurance schemes and rules are used to achieve zero-defect production status and risk-free certification.  A strategy for the integration of a variety of functions matching uniquely assembled machining processes brings astonishing results.

Efficient quality inspection requires the solution of two major problems: one is assurance of accuracy and completeness of paperless data capture and another is intelligent data processing providing predictive control and low-risk quality certification.  The solution leads to a great new area of interrelations between paperless quality inspections, predictive process control, embedded knowledge of mechanics of tool wear, and statistical evidence of zero-defect quality. Benefits in inspection efficiency and effectiveness provided by the MICRONITE system come from properly identifying the details of inspection process and presenting data to automated analysis and decision making. An essential requirement to an inspection process is the provision of right information for any size of production volume. Structurally, inspection feature of MICRONITE consists of (1) operation ID with import of CAD drawing, (2) process-adaptive control plan, (3) paperless and manual inspection data capture on the operator’s workstation and inspector’s  quality station, (4) user-tailored interfaces with action decisions, and (5) quality certificates. The entire inspection process includes in-process inspection, offline (final) inspection, and quality-related capability validation. Quality information from the individual functions is used for overall quality assessment and certification. The fast and easy implementation of MICRONITE’s inspection process helps focus on quality improvement and reduction of total cost quality.

The purpose of this book is to expose engineers to the conceptual ideas and functional characteristics of knowledge-based support of advanced manufacturing, in particular machining and grinding.  The knowledge-based system MICRON ITE  contains integrated components that focus on technology, management, and the workforce. It leads users to well-coordinated data-driven solutions.  There are four strategic principles that are used to define the scope of process solutions using the knowledge-based system.

First, constructive utilization of existent and emerging data capture resources. They usefulness is judged by the ability to clarify positive and negative components influencing overall production effectiveness. Second, in a knowledge-based  environment, elimination of non-productive engineering time is feasible due to an automated analysis of real-time and offline data. Third, interdisciplinary communication between quality and process engineers and production supervisors enhances the results of automated decision making process. Fourth, and finally, information and acquired knowledge is used over time – therefore the strategic impact of knowledge-based solutions  continuously improves the bottom line.

With the current pace of technological change, specialized process control techniques provide the quickest route to achievement of acceptable production effectiveness which is necessary for new product manufacturing. Knowledge-based solutions provide the means of

matching shop floor and engineering data to complexity of modern machining operations.

The purpose of this book is to provide a background  and framework for visualization of CNC machining process utilizing inspection and tooling data. Logistics includes assignment of dimensional characteristics to cutting tools, paperless data capture, multi-feature visualization of tool controllability, automated tool-and-quality data filtering , and automated process actions analysis. No matter how good process design, without effective visualization of CNC machining process, a company cannot achieve peak performance. The long-term goal is to establish a standard methodology for interactive process optimization involving operators and engineers.  There is an absolute need to convert inspection data into a clear visualization of dynamic improvement of both process design and process control. Losses within manufacturing  are reduced by increasingly visible feedback provided by the knowledge-based machining control system.

Several advantages of automated process visualization is presented: (1) accuracy in evaluation of tool and process capability, (2) consistency in representation of thru-operation process models, (3) efficiency in rule-based filtering of tools’ performance, (4) separation of domain knowledge between real-time production data and offline engineering studies, and (5) the ability to share multidisciplinary data and an integrated data set. This makes  the system readily implemented and provides great savings in the time and effort involved in particular applications.

Measurement accuracy and precision is a key problems towards effective control of close and extremely close tolerances. The present document describes an advanced methodology and applied techniques to discover and quantify systematic and random measurement errors. The application of the methodology requires ultimate flexibility of design of metrological studies which ordinary include measurement instruments, observers, and methods. An important element of measurement system acceptance is the need for cluster-organized data manipulation involving in-screen formation of sets of selected variables. Another important component of metrological system validation is an array of  acceptance criteria describing shifts, spread, measurement capability, and permissible tolerance limits. It is shown that with the use of specialized metrology analytics, major causes of measurement errors can be easily discovered and their effect can be substantially reduced.

The software performance and effectiveness  depends on achieving the obvious goal of assuring the absolute accuracy of measurements and non-conventional purpose of attaining the ability to recognize dynamic mechanisms of machining process.  The MICRONITE approach for process control is different from that used by statistical software packages. MICRONITE has three major focuses. First, a variety of models are employed for real-time and off-line data analysis. Assurance of extremely low risk of quality acceptance and process control errors is absolutely necessary to benefit from massive amount of data. Second, MICRONITE addresses the needs of tool-centered control of machining process. Along with grouping of characteristics by finishing tools, the system determine the capability of form and single-point tools to hold multiple tolerances. Third, measuring and machining metrology is included into the knowledge-based control system. There is significant consideration of measuring errors and form variation that would be worst for the ability to correctly certify product quality and provide preventive zero-defect process control. The outcome is a broad-based plan for the best utilization of modern equipment.

The use of the knowledge-based process recognition system provides quick reaction to the variation and change of drift rate with the possibility of early warning and stop decision. In order to consistently produce high accuracy parts MICRONITE’s control methods are designed around process-specific concepts in terms of the type of model, measurement accuracy, and acceptance parameters. Indeed, MICRONITE is equipped with all control tools to achieve extreme dimensional accuracy and surface finish

The purpose of this document is to describe and put into prospective new methodologies and techniques which support cost-efficient prevention of non-conforming and minimum risk of acceptance errors during certification process. A framework identifying three components basic to preventive quality control is developed. These components include: tightly controlled inspection data captured during a production run, automation of acceptance decisions  for random inspection by variables (final and incoming inspection) and capability verification  related to measurement  and machining equipment. It is shown that Inspection data utilization is much more than the creation of reports containing raw data and calculation of indexes Cpk and Ppk. MICRON ITE brings to quality control ability to quickly resolve problems and analyze data with significantly enhanced accuracy. Also, standard operating procedures (SOPs), describing product-related inspection plans and certification process are presented. The typical cases of customer-required SOPs  which include all elements necessary to meet regulations are discussed.

The CMM performance and effectiveness  depends on achieving the obvious goal of assuring the absolute accuracy of measurements and purpose of attaining the ability to recognize and control product quality and dynamics of machining process.  The MICRONITE approach for process control with CMM is different from the one which is used by statistical (SPC) software. MICRONITE has three major focuses. First, a variety of engineering models are developed for real-time and off-line data analysis. Assurance of extremely low risk of quality acceptance and process control errors is absolutely necessary to benefit from massive amounts of CMM data. Second, MICRONITE addresses the needs of tool-centric control of machining process. Along with grouping of characteristics by finishing tools, the system determines the capability of form and single-point tools to hold multiple tolerances. Third, measuring and machining metrology is included into the knowledge-based control system. There is significant consideration of elimination of excessive measuring errors and form variation that would be detrimental for the ability to correctly certify product quality and provide preventive zero-defect process control.

The achievement of potential benefits depends on provision of cross-functional MICRONITE-based communication between process engineering, quality, and production.  MICRONITE’s Artificial Process Manager demonstrates capabilities to perform massive CMM data interpretation to adapt to complex machining processes and provide zero-defect environment with reduced inspection and improved throughput.

Effectiveness of Vision systems has been recognized for its good results on speedy inspection. However, conventional SPC and other statistical tools as well as visual process control interfaces fall short in achievement of accurate predictive control of close-tolerance characteristics  and real-time assessment of cutting tool condition.

MICRONITE  interface for Vision System allows to (1) achieve high-level broad scope utilization of Vision Systems using new class of engineered process control models and statistical estimates applied to operations with perishable tools, (2) widen areas of application for Vision Systems from single samples to predictive tool-and-quality control, risk-free acceptance of product quality, machine-and-tool diagnostics, measurement process validation, and optimization of machining conditions, (3) implement advanced MICRONITE metrology analytics for offline and in-process validation of measurement accuracy and reliability, (4) create diversity of control plan options to succeed in zero-defect quality and superior effectiveness of process control, (5) automate Vision Machine data analysis and the decision making process for control of tool offsets and tool changes, and (6)  integrate automatic data capture between Vision machines, CMMs, and hand-held gages providing real-time intelligent control of the entire operation. Essentially, MICRONITE created machining-specific software suites and components which are successfully used for utilization of Vision System data.

It has long been standard practice for random sampling to use acceptance system STD-105E (ANSI/ASQC Z1.4). Along with large sample sizes, acceptance by attributes bears the high risk to accept a bad lot (Type II error, consumer’s risk).  Acceptance sampling system STD-414 (ANSI/ASQC  Z1.4) allows lesser sample size and reduces Type II errors. For the first time, MICRONITE provides automation of acceptance decisions for MIL-STD-414 enhancing computational  logistics for acceptance by variables

However, the standard system for inspection by variables can be used only for normally distributed sample data and requires historical data for reduced sample sizes. MICRONITE introduces new class of acceptance systems  with the wide array of sampling plans, standard and proprietary statistical estimates of population min/max and deviation, rule-based acceptance decisions and automated sample size control. MICRONITE’s sampling control is a valuable tool for final and receiving inspection, and efficient verification of process data.

As a result, MICRONITE is able to: (1) provide reliable prediction of non-conformance in random sampling by variables, (2)assure precise estimation of deviation from specification limits for an entire production lot using relatively small sample size, (3) document product quality with high-accuracy statistical reports, (4) minimize sample size maintaining low risk of acceptance errors, (5) reduce labor and time for lot ID, control plan design, and documentation, (6) provide paperless traceability of inspection records, and (7) replace ineffective system for inspection by attributes with advanced systems for inspection by variables

The  main topics related to control of small-batch machining are: (1) automated operation ID including import of CAD drawing, (2) paperless data acquisition and statistical data representation by single piece or stream of pieces forming samples of one, two, or more, (3) optional assignment of dimensional characteristics to cutting tool, (4) AI-based analysis of trend and variation with tool wear simulation , and (5) series of product certification reports.

Process recognition and predictive control for sample size of one includes several phases. In a first phase, location of reading (out-of-spec, in RED/YELOW zone) determines the action decision.  In a second phase, trend-and-variation parameters are extracted from location of model-connected single pieces. These parameters are re-computed after each piece has been measured. In a third phase, relations between computed parameters are estimated and prioritized.. Finally, a rule-based backward model makes the action decision based on the results of trend-and-variation analysis and severity ranking. Indexes CP 90 and Cpk 90 (90% confidence level) are used for assessment of process capability

This document describes a conceptual model which integrates static process capability and dynamic process controllability  in details how to conduct process capability studies along with process controllability verification. It defines a standard methodology for the purpose of characterizing and optimizing machines, gauges, equipment and manufacturing processes. The methodology goes beyond just determining capability and stability. It also leads to identifying and reducing the major sources of variability, that is, optimizing the equipment or process and then establishing a total proactive control to eliminate the possibility of future defects. The methodology is presented and divided into six elements: (1) process characterization, (2) analytics of machine and measurement metrology, (3) determination of process capability and controllability, (5) optimization of critical parameters, and (6) real-time monitoring of vital capability parameters. The six elements present a logical connection and designed to preserve and guarantee statistical accuracy throughout the analysis. Short-term and long-term capability study approaches are explained and their respective methodologies are discussed. Standard forms and worksheets are used throughout the document to lead the user through all important steps necessary to achieve capable manufacturing processes.

The purpose of machining-specific control plan engineering is to ensure that an inspection schemes effectively provide process recognition and predictive control, quality certification, and cover required engineering support functions. This task is complicated and promising enough to warrant the use of computerized control planning, with following four roles: (1) identifying type and inspection level of characteristic-adaptive control models, (2) identifying mode and adaptive/fixed control of sampling interval, (3) identifying a plan for final inspection and quality audit, and (4) identifying features of metrological and capability studies. The achievement of plan effectiveness hinges on two activities: recording the accurate and complete inspection data and converting the collected data into helpful and productive information. Information flow paths to and from MICRONITE embrace the entire plant. Feedback from the production process into the control plan optimization stage is considered a final aim in achievement of near-perfect control plan design. An extensive description of control plans for CNC Swiss, Machining Centers, Lathes, Multi-spindle automatics, and Hydromats  are presented.

This document discloses unique features of an hierarchical production control system for machine shops as an alternative to generic concept of O.E.E. A production control feature of MICRONITE system is linked to process / tool control features capable of processing data for different jobs simultaneously. Its effective use in a particular situation is largely determined by its degree of flexibility and multi-functionality. A well-designed data capture interface is a key condition to deal properly with the specific plant environment..  Plant-wide dashboard s allow to know the underlying causes of production gaps and close those gaps during the production day. Users can drill down from any dashboard or report to the specific causes and corrective actions. The right information for the process, quality, and production at the time being performed keeps everyone aligned to work-in-progress. The feature contains technical work instructions, proactive maintenance actions, standard operating procedures. Visual control panels present current production rates, production delays, and other valuable information. Contrasting to O.E.E. the user can modify formulas for calculation of true production effectiveness tailored to particular plant environment.

The MICRONITE system is designed to focus a company on a new concept of Management By Priorities (MBP). MBP is most suitable for computer-integrated process and production control because of its capability to bind together static and dynamic components of production. machining. Management by priorities is fundamentally different from MES (Manufacturing Execution System)  and Quality Management Systems in that it emphasizes automated problem prioritization within the complete production chain from in-process inspection to tool performance to production efficiency and overall time loss analysis. MBP is a response to machining complexity  brought about by tightened quality requirements for close tolerances and competitive cost pressures. Prioritization is automatically done to determine if a process is capable (static component) and controllable (dynamic component), inspection and action discipline is supervised, non-productive time is identified and reported, and production effectiveness is sufficiently manageable. MICRONITE MBP analyses data in both Top-Down and Down-TOP directions providing a unique value for a customer and profit-focused machined shop.

There are a number of important points to be realized in relation to design of computerized process control for the use by a machinist. It is extremely unlikely that a machinist will embrace the system which requires a substantial time for data capture and does not return to a machinist a helpful feedback. Considerable resistance is expected to what machinists perceive as being forced to do with little or no benefits. This resistance can be overcome, in view of the system which is knowledgeable in machining and capable of making right decisions. Except for process control decisions, a machinist is looking for other essential functions, such as alignment of dimensional characteristics created by the same tool, part-specific capabilities to hold tight tolerance, data-led decisions for proactive maintenance, and problem solving through real-time communication with supervisors and engineers. Defining the link between machinist’s responsibilities and completeness of machining control system is one of the most important steps of successful system implementation.

In reality, a developer should ask: who is responsible for the result? – Is it up to the shop floor to implement the system, or it is up to system developers to go to such details and perfection that it is obvious to implement it? The answer lies in a marriage of shop floor dedication and absolute fitness of control functions to production environment. Once this  task is accomplished, the progress will be continually achieved.