Michael P. Brundage

An Overview of a Smart Manufacturing System Readiness Assessment

Smart manufacturing, today, is the ability to continuously maintain and improve performance, with intensive use of information, in response to the changing environments. Technologies for creating smart manufacturing systems or factories are becoming increasingly abundant. Consequently, manufacturers, large and small, need to correctly select and prioritize these technologies correctly. In addition, other improvements may be necessary to receive the greatest benefit from the selected technology. This paper proposes a method for assessing a factory for its readiness to implement those technologies. The proposed readiness levels provide users with an indication of their current factory state when compared against a reference model. Knowing this state, users can develop a plan to increase their readiness. Through validation analysis, we show that the assessment has a positive correlation with the operational performance.

Smart manufacturing through a framework for a knowledge-based diagnosis system

Various techniques are used to diagnose problems throughout all levels of the organization within the manufacturing industry. Often times, this root cause analysis is ad-hoc with no standard representation for artifacts or terminology (i.e., no standard representation for terms used in techniques such as fishbone diagrams, 5 why’s, etc.). Once a problem is diagnosed and alleviated, the results are discarded or stored locally as paper/digital text documents. When the same or similar problem reoccurs with different employees or in a different factory, the whole process has to be repeated without taking advantage of knowledge gained from previous problem(s) and corresponding solution(s). When discussing the diagnosis, personnel may miscommunicate over terms used in the root cause analysis leading to wasted time and errors. This paper presents a framework for a knowledge-based manufacturing diagnosis system that aims to alleviate these miscommunications. By learning from diagnosis methods used in manufacturing and in the medical community, this paper proposes a framework which integrates and formalizes root cause analysis by categorizing faults and failures that span multiple organizational levels. The proposed framework aims to enable manufacturing operations by leveraging machine learning and semantic technologies for the manufacturing system diagnosis. A use case for the manufacture of a bottle opener demonstrates the framework. INTRODUCTION Root cause analysis is used in many industries to find the causes of different faults in a system to provide corrective and preventative action (CAPA) plans to alleviate those faults [1-5]. Traditionally, root cause analysis methods in the manufacturing 1 Corresponding author: mpb1@nist.gov industry do not present themselves for formal retrieval of information from past studies: it is a one-off practice [1,6-7]. Often, crucial elements of the problem solving process are paperborne techniques, which do not lend themselves to automated storage and retrieval. The Quality Information Framework (QIF) is beginning to formalize cause and effect retrieval, however it is still free-form text-based, with no formal schema defined [8]. This paper presents a framework for a knowledge-based system for root cause analysis. The knowledge-based method provides a more formal structure for manufacturing system diagnosis. By synthesizing different approaches from engineering and the medical community, this framework allows for more accurate communication, discovery, and reuse of manufacturing diagnosis and corrective and preventative action plans. By providing more structure and learning from the data, manufacturers can reduce the number of misdiagnoses and decrease time to investigate issues. Currently, the medical industry has a more formal diagnosis procedure than in the manufacturing industry with more research into knowledge-based methods [9-10]. Symptom checker websites, such as ‘symcat.com’ and ‘symptoms.webmd.com,’ allow users to input symptoms and output corresponding diseases listed with the probability of occurrence. The goal of this framework is to apply techniques learned from these symptom checker and medical diagnosis systems to provide a manufacturing diagnosis system. There are four main concepts of medical diagnosis in the medical community [10-11]:  Problem assessment is an evaluation of the patient to assess the current condition,

Analyzing environmental sustainability methods for use earlier in the product lifecycle

Environmental sustainability information in the manufacturing industry is not easily shared between stages in the product lifecycle. In particular, reliable manufacturing-related information for assessing the sustainability of a product is often unavailable at the design stage. Instead, designers rely on aggregated, often outdated information or make decisions by analogy (e.g., a similar manufacturing process for a similar product yielded X and Y results). However, smart manufacturing and the Internet of Things have potential to bridge the gap between design and manufacturing through data and knowledge sharing. This paper analyzes environmental sustainability assessment methods to enable more accurate decisions earlier in design. The techniques and methods are categorized based on the stage they apply to in the product lifecycle, as described by the Systems Integration of Manufacturing Applications (SIMA) reference architecture. Furthermore, opportunities for aligning standard data representation to promote sustainability assessment during design are identified.

A Review of Current Machine Learning Techniques Used in Manufacturing Diagnosis

Artificial intelligence applications are increasing due to advances in data collection systems, algorithms, and affordability of computing power. Within the manufacturing industry, machine learning algorithms are often used for improving manufacturing system fault diagnosis. This study focuses on a review of recent fault diagnosis applications in manufacturing that are based on several prominent machine learning algorithms. Papers published from 2007 to 2017 were reviewed and keywords were used to identify 20 articles spanning the most prominent machine learning algorithms. Most articles reviewed consisted of training data obtained from sensors attached to the equipment. The training of the machine learning algorithm consisted of designed experiments to simulate different faulty and normal processing conditions. The areas of application varied from wear of cutting tool in computer numeric control (CNC) machine, surface roughness fault, to wafer etching process in semiconductor manufacturing. In all cases, high fault classification rates were obtained. As the interest in smart manufacturing increases, this review serves to address one of the cornerstones of emerging production systems.

Procedure for Selecting Key Performance Indicators for Sustainable Manufacturing

The need for an open, inclusive, and neutral procedure in selecting key performance indicators (KPIs) for sustainable manufacturing has been increasing. The reason is that manufacturers seek to determine what to measure to improve environmental sustainability of their products and manufacturing processes. A difficulty arises in understanding and selecting specific indicators from many stand-alone indicator sets available. This paper presents a procedure for individual manufacturers to select KPIs for measuring, monitoring and improving environmental aspects of manufacturing processes. The procedure is the basis for a guideline, being proposed for standardization within ASTM International. That guide can be used for (1) identifying candidate KPIs from existing sources, (2) defining new candidate KPIs, (3) selecting 1 Corresponding author: shaw.feng@nist.gov ASME Journal of Manufacturing Science and Engineering 2 appropriate KPIs based on KPI criteria, and (4) composing the selected KPIs with assigned weights into a set. The paper explains how the developed procedure complements existing indicator sets and sustainability-measurement approaches at the manufacturing process level.

Maturity Models and Tools for Enabling Smart Manufacturing Systems: Comparison and Reflections for Future Developments

One of the most exciting new capabilities in Smart Manufacturing (SM) and Cyber-Physical Production Systems (CPPS) is the provisioning of manufacturing services as unbundled “apps or services”, which could be significantly more flexible and less expensive to use than the current generation of monolithic manufacturing applications. However, bundling and integrating heterogeneous services in the form of such apps or composite services is not a trivial job. There is a need for service vendors, cloud vendors, manufacturers, and other stakeholders to work collaboratively to simplify the effort to “mix-and-match” and compose the apps or services. In this regard, a workshop was organized by the National Institute of Standards and Technology (NIST) and the Open Applications Group Inc. (OAGi), with the purpose to identify – through parallel sessions – technology and standard needs for improving interoperability and composability between services. The workshop was organized into five working session. This paper documents evidences gathered during the “Smart Manufacturing Systems Characterization” (SMSC) session, which aims at establishing a roadmap for a unified framework for assessing a manufacturer’s capability, maturity and readiness level to implement Smart Manufacturing. To that end, the technology maturity, information connectivity maturity, process maturity, organizational maturity, and personnel capability and maturity, have been identified as critical aspects for Smart Manufacturing adoptions. The workshop session culminated at providing a coherent model and method for assisting manufacturing companies in their journey to smart manufacturing realizations. This paper shows three different maturity models and tools that, thanks to their complementarity, enable one to reflect on the different perspectives required by SMSC. These models and tools are usable together for assessing a manufacturing company’s ability to initiate the digital transformation of its processes towards Smart Manufacturing. Therefore, based on their comparison, the ultimate purpose of the research is to come up with a set of coherent guidelines for assessing a manufacturing system and its management practices for identifying improvement opportunities and for recommending SM technologies and standards for adoption by manufacturers.

Toward Semi-autonomous Information

To facilitate root cause analysis in the manufacturing industry, maintenance technicians often fill out “maintenance tickets” to track issues and corresponding corrective actions. A database of these maintenance-logs can provide problem descriptions, causes, and treatments for the facility at large. However, when similar issues occur, different technicians rarely describe the same problem in an identical manner. This leads to description inconsistencies within the database, which makes it difficult to categorize issues or learn from similar cause-effect relationships. If such relationships could be identified, there is the potential to discover more insight into system performance. One way to address this opportunity is via the application of natural language processing (NLP) techniques to tag similar ticket descriptions, allowing for more formalized statistical learning of patterns in the maintenance data as a special type of short-text data. This paper showcases a proof-of-concept pipeline for merging multiple machine learning (ML) and NLP techniques to cluster and tag maintenance data, as part of a broader research thrust to extract insight from largely unstructured natural-language maintenance logs. The accuracy of the proposed method is tested on real data from a small manufacturer.