AI Glossary

This glossary of artificial intelligence terms and concepts is designed to help you navigate the world of industrial AI, machine learning, and beyond.

Reset all filters
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Adaptive Synthetic Sampling (ADASYN)

An oversampling technique that is used to handle imbalanced datasets. It adapts the density distribution of the data by creating more synthetic data in the region of the feature space where the density of minority class examples is low, and fewer or none where the density is high.


In reinforcement learning, an agent is an entity that interacts with an environment, observes its current state, and takes actions to maximize some notion of cumulative reward.


A set of instructions designed to perform a specific task or solve a particular problem.

Artificial General Intelligence (AGI)

AI that is capable of performing any intellectual task that a human can.

Artificial Intelligence (AI)

The ability of machines to perform tasks that would normally require human intelligence.

Asset Performance Management (APM)

A comprehensive approach to maintaining and improving the performance of physical assets, such as machinery, equipment, and infrastructure.


The ability of a system or machine to operate independently, without human intervention or control.

Bayesian Inference

A statistical method for updating beliefs or probabilities based on new evidence or data.

Big Data

Extremely large datasets that can be analyzed to reveal patterns, trends, and associations.

Carbon Dioxide Equivalent (CO2e)

A measure of the amount of greenhouse gases (GHGs) emitted as a result of human activities, expressed in terms of the amount of carbon dioxide (CO2) that would have the same global warming potential over a specified time period.


An AI-powered program that simulates conversation with human users.


In machine learning, classification refers to a type of supervised machine learning technique that involves predicting a categorical or discrete output variable based on one or more input variables. The goal of classification is to build a model that can accurately classify new data points based on the patterns learned from the training data.

Closed-Loop Control

A control policy that uses feedback from the system being controlled to adjust or modify the control inputs in real-time.

Computational Learning Theory

A branch of computer science and artificial intelligence that focuses on the design and analysis of algorithms and models for machine learning.

Computer Science and Artificial Intelligence Laboratory (CSAIL)

A research laboratory at the Massachusetts Institute of Technology (MIT) dedicated to advancing the fields of computer science and artificial intelligence.

Computer Vision (CV)

The ability of computers to interpret and analyze images and video.

Computerized Maintenance Management System (CMMS)

A software system that helps organizations manage and track maintenance activities for their assets and equipment. CMMS is used in industries such as manufacturing, utilities, facilities management, defense, and transportation.


In general, a control is something that regulates or directs the behavior or operation of a system or process. In the context of engineering and technology, control refers specifically to the management or regulation of a system or process to achieve a desired outcome or to optimize performance.

Control Optimization

The process of finding the best possible control policy for a given system, in order to achieve a desired outcome or to optimize performance.

Control Policies

The strategies or methods used to control or manage a system or process. In the context of engineering, control policies are used to manage or regulate a system or process to achieve a desired outcome or to optimize performance.

Cybersecurity Maturity Model Certification (CCMC)

A framework created by the U.S. Department of Defense (DoD) to assess and enhance the cybersecurity posture of companies that do business with the DoD.

Data Augmentation

A strategy that allows significantly increasing the diversity of data available for training models, without actually collecting new data. This provides a way to add variability and improve the ability of models to generalize, leading to improved model performance.

Data Mining

The process of discovering patterns and insights in large datasets.

Data Pipeline

A series of processes or systems that are used to collect, process, and transform data from various sources and move it to a destination where it can be used for analysis or other applications.

Data Science

An interdisciplinary field that involves the use of statistical, computational, and other quantitative methods to extract insights and knowledge from data.

Data Validation Rules

Specific guidelines established to ensure the quality, accuracy, and consistency of the data entering a system.

Decision Trees

A popular type of supervised learning algorithm used for both classification and regression tasks. They are based on a tree-like model of decisions and their possible consequences.

Deep Learning (DL)

A type of machine learning that uses artificial neural networks to learn from data and make predictions.

Dissolved Gas Analysis (DGA)

A technique used to detect and diagnose faults in oil-insulated electrical equipment, such as transformers and circuit breakers.

Distributed Control System (DCS)

A computerized control system used to monitor and control industrial processes, such as those in HVAC, power plants, chemical plants, and manufacturing facilities.

Engineering, Procurement, and Construction (EPC)

A project delivery method used in the construction industry. EPC contracts are commonly used in the construction of large-scale infrastructure projects, such as power plants, oil and gas facilities, and transportation systems.

Enterprise Resource Planning (ERP)

A business management system that integrates various processes and functions (i.e., finance, human resources, manufacturing) across an organization into a single, centralized system.

Expert Systems

AI systems that mimic the decision-making abilities of a human expert in a specific field.


The process of estimating values for input variables that lie outside the range of observed data points. It involves making predictions or forecasting for values that are beyond the scope of the training data.

Federated Learning

A machine learning approach where a model is trained across multiple decentralized devices or servers each holding local data samples, without exchanging them.

Gaussian Process (GP)

A probabilistic model commonly used for regression and classification tasks. It is a non-parametric, flexible approach to modeling complex functions that can be used for tasks such as interpolation, extrapolation, and uncertainty quantification.

Geographic Information System (GIS)

A computer-based tool for storing, analyzing, and visualizing geospatial data. GIS allows users to capture, manage, and display data related to geographic locations, such as land use, topography, and population.

Heating, Ventilation, and Air Conditioning (HVAC)

Systems used to provide indoor comfort and air quality in residential, commercial, and industrial buildings.

Hidden Markov Model (HMM)

A statistical model used to model temporal data with an underlying hidden structure. HMM is a type of probabilistic graphical model that allows for the analysis and modeling of sequences of data, such as speech signals, biological sequences, or financial data.

Incremental Learning

A machine learning approach where the model learns continuously, updating its knowledge as new data comes in.

Industrial Internet of Things (IIoT)

The use of Internet of Things (IoT) technologies and concepts in industrial settings, such as manufacturing, logistics, and energy production.

Instrumentation and Controls Engineer (ICE)

A professional responsible for designing, developing, and maintaining systems that control and monitor various processes and equipment in industrial settings. These systems may involve the use of sensors, actuators, controllers, and other devices to automate and optimize processes.

Internet of Things (IoT)

A network of physical devices connected to the internet, which can be monitored and controlled remotely using AI.


In machine learning, interpolation refers to the process of estimating values for points within a range of observed data points.

Levelized Cost of Energy (LCOE)

A metric used to evaluate the cost of producing energy from a particular source over its lifetime, expressed as the cost of producing one unit of energy (e.g. one kilowatt-hour or one megawatt-hour).

Logistics Response Time (LRT)

An important metric for logistics and supply chain management, LRT refers to the amount of time it takes for a logistics system to respond to a request for goods or services.

Machine Learning (ML)

A subset of AI that involves training algorithms to make predictions or decisions based on data.

Machine Learning Operations (MLOps)

A set of best practices and tools used to streamline and automate the process of deploying, managing, and monitoring machine learning models in production environments.

Maintenance Cycle Time (MCT)

A metric used to measure the amount of time it takes to complete a maintenance task or cycle, from the moment a request is made until the task is completed.

Mean Time To Failure (MTTF)

A reliability metric that is used to estimate the expected time between the start of operation of a system or component and the occurrence of its first failure.


A mathematical framework or algorithm that is designed to learn from data and make predictions or decisions based on that data.

Natural Language Generation (NLG)

The process of automatically producing human-like text from data or structured information.

Natural Language Processing (NLP)

The ability of computers to understand and analyze human language.

Neural Network

A type of algorithm modeled after the structure of the human brain, used in deep learning.

Open-Loop Control

A control policy that does not use feedback from the system being controlled. Instead, it relies on a predetermined set of inputs or commands to achieve a desired output.

Operating Mode (OM)

The specific state or condition of an operating system, machine, or device at a particular point in time.

Operations & Maintenance (O&M)

The activities and processes involved in managing and maintaining a physical asset, such as a building, a machine, or an infrastructure system.


A common problem in machine learning and artificial intelligence, where a model performs well on the training data but poorly on unseen data (like validation or test data).

Physics-based Modeling

A type of computer simulation that relies on the laws of physics to predict or analyze real-world phenomena. Unlike data-driven models, which rely on analyzing large amounts of data to make predictions, physics-based models use established physical laws to predict outcomes.


In reinforcement learning, a policy refers to the strategy or rule that an agent uses to determine its actions in a given state.


In machine learning, prediction refers to the process of using a trained model to make predictions or forecasts about future or unseen data.

Predictive Analytics

The process of using data, statistical algorithms, and machine learning techniques to identify the likelihood of future outcomes based on historical data.

Programmable Logic Controller (PLC)

A type of digital computer that is designed to control and automate industrial processes, machinery, and equipment.

Radio Frequency Identification (RFID)

A technology that uses radio waves to identify and track objects, people, or animals.


A type of supervised machine learning technique used for predicting continuous numerical output values based on one or more input variables.

Reinforcement Learning

A type of ML in which an agent learns to make decisions by receiving rewards or punishments and using that information to alter its decision-making processes.

Remaining Useful Life (RUL)

Used in predictive maintenance, RUL is the estimated time period that a piece of equipment or system can continue to operate before it reaches the end of its useful life.

Remote Monitoring System (RMS)

Enables remote operators and managers to monitor and control equipment and processes from a centralized location, reducing the need for on-site inspections and maintenance.


The design, construction, and operation of robots to perform tasks.

Run To Failure (RTF)

A maintenance strategy in which equipment or assets are operated until they fail, without any preventive or proactive maintenance.

Semi-supervised Operating Mode

A type of machine learning approach where a combination of labeled and unlabeled data is used to train a model.

Sentiment Analysis

The process of using NLP to analyze and understand the emotions and opinions expressed in text.

Set Point Control

A type of control system that is used in many industrial and engineering applications to regulate a process variable, such as temperature, pressure, or flow rate.

Supervisory Control and Data Acquisition (SCADA)

A computer-based system used to monitor and control industrial processes and infrastructure, such as power generation, water treatment, and oil and gas pipelines.

Synthetic Minority Over-sampling Technique (SMOTE)

A machine learning technique used to handle class imbalance in a dataset. It's often used in scenarios where the number of instances of one class far outweighs the instances of the other class(es).

System Average Interruption Duration Index (SAIDI)

A reliability indicator used to measure the average outage duration of an electrical power distribution system.

System Average Interruption Frequency Index (SAIFI)

A performance metric used to measure the reliability of an electric power distribution system, based on how many outages a customer experiences in a certain time period.

Time Series Data

Data that varies over time, often in a continuous or sequential manner, i.e. weather patterns or stock market prices.

Time To Event (TTE)

The duration of time between a specific starting point and a particular event of interest. The event can be any significant occurrence or milestone, such as the failure of a machine, the occurrence of a disease, or the completion of a project.

Training Data

The set of data that is used to train a machine learning model, with the goal of teaching the machine learning algorithm to identify patterns and relationships in the data that can be used to make accurate predictions or classifications for new, unseen data.

Transfer Learning

A machine learning technique where a model developed for one task is reused as the starting point for a model on a second task. It is a popular approach in deep learning because it can train deep neural networks with comparatively little data.

Universal Time Coordinated (UTC)

A global standard for measuring time that is based on the rotation of the Earth. It is used to synchronize clocks and timekeeping systems around the world, allowing for consistent time references across different time zones and regions. UTC is widely used in various fields, including telecommunications, aviation, finance, and computer systems.

Value Stream Mapping (VSM)

A lean manufacturing tool used to identify inefficiencies and waste in the production process and to develop strategies to improve overall efficiency and reduce costs. Value stream mapping involves visually mapping out the steps and flow of materials, information, and activities involved in delivering a product or service. It helps identify areas of improvement, such as bottlenecks, excessive waiting times, unnecessary tasks, and non-value-added activities, with the aim of streamlining processes and optimizing value creation.

Work Management System (WMS)

A software system designed to manage and track the progress of work-related activities within an organization. WMS typically provides functionalities for planning, scheduling, assigning tasks, monitoring progress, and generating reports related to work assignments and resources. It helps streamline workflow, enhance collaboration among team members, optimize resource allocation, and improve overall productivity and efficiency.

No results were found based on your search!
Try a different keyword or clear your input
Clear Search Input