Clarifying Data Analytics Concepts for Industrial Engineering

Clarifying Data Analytics Concepts for Industrial Engineering

1. Research in data analysis for smart manufacturing has become a mandatory trend

   • Data analytics wants to analyse raw data in order to discover hidden patterns and relationships among different variables.
     • the main aim of data analysis is to extract useful information from rough data and transfer it to effective knowledge to improve product and process understanding and to support decisions (Ge et al. 2017
       • i.e. To send the right information to the right person in the right moment, which is boosted by the digitization trends that are permeating the manufacturing industry context

2. Vocabulary

   • Data mining is the extraction of implicit, previously unknown and potentially useful information from data. The idea is to construct numerical algorithms that investigate datasets to extract hidden patterns and regularities. Patterns that reveal strong structures can be used to predict future outcomes.
     • Categories
       • 1. Supervised learning deals with predicting an outcome value based on the measures of other input variables.
       • 2. Unsupervised learning deals with description of associations and patterns among a set of input data. In this case, we observe only the features and we are not able to define an output. The final aim is to find hidden patterns, to describe data and to find out how they may be organized or clustered
     • Through data mining it is possible to achieve competitive advantages and the superiority of data mining and machine learning over other experimental techniques lies in the fact that data can be collected during the normal operations of the manufacturing process (Harding et al. 2006).
   • Machine learning is the field that developed most of the algorithms to find and describe structural patterns in data. So, we can say that machine learning represents the technology instruments of data mining.
     • TECHNIQUES IN SMART MANUFACTURING
       • Supervised Learning
       • Linear Methods for Regression
         • simple and often provide an adequate and interpretable description of how the inputs affect the output
           • A linear regression model assumes that the regression function is linear in the variable inputs X1, … Xp. Multivariate linear regression is a generalization of linear regression, by considering more than one dependent variable
       • Linear Method for Classification
         • The linear classifier uses the variable’s characteristics to identify which class (or group) the variable belongs to.
       • Decision Tree Based Method
         • . Nodes in a decision tree involve testing a particular attribute.
       • Neural Networks
         • The central idea is to extract linear combinations of the inputs as new features, and then model the target output as a nonlinear function of these features.
         • The Artificial Neural Networks (ANNs) are based on the human brain by making the right connections between neurons. It is a flexible mathematical structure and the structure is capable of identifying complex nonlinear relationships between input and output datasets
       • Unsupervised Learning
       • Principal Component Analysis
         • converts a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables
       • Association Rules
         • Its goal is to find joint values of the input variables that appear most frequently in the dataset, i.e. to predict combinations of attributes.
       • Cluster Analysis
         • used to recognize whether the dataset consists of distinct subgroups, each group representing objects with substantially different properties.
       • Self-Organizing Maps (SOM)
         • observations can be mapped down onto the two-dimensional coordinate syste
   • Data Analytics: composed of: initial standard statistical techniques, such as visualization of data, selection of attributes, discarding outliers etc. in order to construct the initial example set; then, machine learning algorithms construct rules and statistical tests are used to validate them
   • AI: “branch of computer science devoted to developing data processing systems that perform functions normally associated with human intelligence, such as reasoning, learning, and self-improvement”.
   • Big Data refer to a type of datasets, in which variables are different in type and nature, for this reason, they compel to be analysed with a number of specific techniques, models and algorithms, that belong to the Data Analytics, Data Mining and Machine Learning