0000000001157376
AUTHOR
David Lechevalier
showing 11 related works from this author
A methodology for the semi-automatic generation of analytical models in manufacturing
2018
International audience; Advanced analytics can enable manufacturing engineers to improve product quality and achieve equipment and resource efficiency gains using large amounts of data collected during manufacturing. Manufacturing engineers, however, often lack the expertise to apply advanced analytics, relying instead on frequent consultations with data scientists. Furthermore, collaborations between manufacturing engineers and data scientists have resulted in highly specialized applications that are not relevant to broader use cases. The manufacturing industry can benefit from the techniques applied in these collaborations if they can be generalized for a wide range of manufacturing probl…
Predictive Model Markup Language (PMML) Representation of Bayesian Networks: An Application in Manufacturing
2018
International audience; Bayesian networks (BNs) represent a promising approach for the aggregation of multiple uncertainty sources in manufacturing networks and other engineering systems for the purposes of uncertainty quantification, risk analysis, and quality control. A standardized representation for BN models will aid in their communication and exchange across the web. This article presents an extension to the predictive model markup language (PMML) standard for the representation of a BN, which may consist of discrete variables, continuous variables, or their combination. The PMML standard is based on extensible markup language (XML) and used for the representation of analytical models…
A Virtual Milling Machine Model to Generate Machine-Monitoring Data for Predictive Analytics
2015
Real data from manufacturing processes are essential to create useful insights for decision-making. However, acquiring real manufacturing data can be expensive and time consuming. To address this issue, we implement a virtual milling machine model to generate machine monitoring data from process plans. MTConnect is used to report the monitoring data. This paper presents (1) the characteristics and specification of milling machine tools, (2) the architecture for implementing the virtual milling machine model, and (3) the integration with a simulation environment for extending to a virtual shop floor model. This paper also includes a case study to explain how to use the virtual milling machin…
Automated uncertainty quantification analysis using a system model and data
2015
International audience; Understanding the sources of, and quantifying the magnitude of, uncertainty can improve decision-making and, thereby, make manufacturing systems more efficient. Achieving this goal requires knowledge in two separate domains: data science and manufacturing. In this paper, we focus on quantifying uncertainty, usually called uncertainty quantification (UQ). More specifically, we propose a methodology to perform UQ automatically using Bayesian networks (BN) constructed from three types of sources: a descriptive system model, physics-based mathematical models, and data. The system model is a high-level model describing the system and its parameters; we develop this model …
Model-based Engineering for the Integration of Manufacturing Systems with Advanced Analytics
2016
To employ data analytics effectively and efficiently on manufacturing systems, engineers and data scientists need to collaborate closely to bring their domain knowledge together. In this paper, we introduce a domain-specific modeling approach to integrate a manufacturing system model with advanced analytics, in particular neural networks, to model predictions. Our approach combines a set of meta-models and transformation rules based on the domain knowledge of manufacturing engineers and data scientists. Our approach uses a model of a manufacturing process and its associated data as inputs, and generates a trained neural network model as an output to predict a quantity of interest. This pape…
Issues in synthetic data generation for advanced manufacturing
2017
To have any chance of application in real world, advanced manufacturing research in data analytics needs to explore and prove itself with real-world manufacturing data. Limited access to real-world data largely contrasts with the need for data of varied types and larger quantity for research. Use of virtual data is a promising approach to make up for the lack of access. This paper explores the issues, identifies challenges, and suggests requirements and desirable features in the generation of virtual data. These issues, requirements, and features can be used by researchers to build virtual data generators and gain experience that will provide data to data scientists while avoiding known or …
Gaussian Process Regression (GPR) Representation in Predictive Model Markup Language (PMML)
2017
International audience; This paper describes Gaussian process regression (GPR) models presented in predictive model markup language (PMML). PMML is an extensible-markup-language (XML) -based standard language used to represent data-mining and predictive analytic models, as well as pre- and post-processed data. The previous PMML version, PMML 4.2, did not provide capabilities for representing probabilistic (stochastic) machine-learning algorithms that are widely used for constructing predictive models taking the associated uncertainties into consideration. The newly released PMML version 4.3, which includes the GPR model, provides new features: confidence bounds and distribution for the pred…
Automated Uncertainty Quantification Through Information Fusion in Manufacturing Processes
2017
International audience; Evaluation of key performance indicators (KPIs) such as energy consumption is essential for decision-making during the design and operation of smart manufacturing systems. The measurements of KPIs are strongly affected by several uncertainty sources such as input material uncertainty, the inherent variability in the manufacturing process, model uncertainty, and the uncertainty in the sensor measurements of operational data. A comprehensive understanding of the uncertainty sources and their effect on the KPIs is required to make the manufacturing processes more efficient. Towards this objective, this paper proposed an automated methodology to generate a hierarchical B…
A Neural Network Meta-Model and its Application for Manufacturing
2015
International audience; Manufacturing generates a vast amount of data both from operations and simulation. Extracting appropriate information from this data can provide insights to increase a manufacturer's competitive advantage through improved sustainability, productivity, and flexibility of their operations. Manufacturers, as well as other industries, have successfully applied a promising statistical learning technique, called neural networks (NNs), to extract meaningful information from large data sets, so called big data. However, the application of NN to manufacturing problems remains limited because it involves the specialized skills of a data scientist. This paper introduces an appr…
Towards a virtual factory prototype
2015
International audience; A virtual factory should represent most of the features and operations of the corresponding real factory. Some of the key features of the virtual factory include the ability to assess performance at multiple resolutions and generate analytics data similar to what is possible in a real factory. One should be able to look at the overall factory performance and be able to drill down to a machine and analyze its performance. It will require a large amount of effort and expertise to build such a virtual factory. This paper describes an effort to build a multiple resolution model of a manufacturing cell. The model provides the ability to study the performance at the cell l…
TOWARDS SMART MANUFACTURING WITH VIRTUAL FACTORY AND DATA ANALYTICS
2017
International audience; Virtual factory models can help improve manufacturing decision making when augmented with data analytics applications. Virtual factory models provide the capability of simulating real factories and generating realistic data streams at the desired level of resolution. Deeper insights can be gained and underlying relationships quantified by channeling the simulation output data to an external analytics tool. This paper describes integration of a virtual factory prototype with a neural network analytics application. The combined capability is used to create a neural network capable of predicting the expected cycle times for a small job shop. The capability can adapt by …