The five facets of a factory: Manpower, Machine, Management, Material, and Method.
AIAMFG
Oracle Adaptive Intelligent Apps for ManuFacturinG.
algorithm
An algorithm is a set of steps to accomplish a task. AIAMFG uses algorithms to process information in a dataset. The algorithm you choose for a model processes information from the dataset (the input source) and compares the results (the output) to a target measure.
API
An Application Programming Interface enables one application to communicate with another application. There are different types of APIs, but AIAMFG uses REST APIs, a type of web service API, which enables two machines to communicate with each other over a network.
Apriori
An algorithm you can select when creating an Insight model.
artificial intelligence
A term used to describe when computers mimic cognitive functions, such as learning and problem solving.
attribute
An attribute describes a data object. For example, a work order (the data object) has a completion date (the attribute). The rows in a database correspond to data objects and the columns correspond to attributes.
bin
An attribute value range. Model target measure results are placed in bins. For example, if a model predicts a quality target measure, such as pH, the bins might be:
Below Lower Limit: 0-7
On Target: 7-8
Above Upper Limit: 8-14
Each bin represents a classification range. Each target measure result is placed into a classification range (bin).
boxplot
A boxplot graph has a box from the first quartile to the third quartile of data. The vertical line through the box represents the median. The whiskers stretch from the first and third quartile to the minimum and maximum.
business entity data
Business entities represent individual instances of data objects, such as customer 1376 (the instance or business entity) for the customer data object. In a Customer database table, each row contains the business entity data for each customer.
case record data
Case record data CSV files capture completed and in-progress work order data in a flattened file format. Each row in the spreadsheet corresponds to one work order and contains all related entity data such as item, operation, routing, and so on, in the same row.
case record identifier
Similar to a key field in a database. In a case record CSV file, the case record identifier, or key field, is a either a work order number or a work order/serial number combination.
Chi-Square
An algorithm you can select when creating a feature significance model.
classification
See: bin.
cloud service
Any software service made available through the internet.
configuration
In AIAMFG, configuration refers to the process of setting up custom data, such as key performance indicators and time series feature sets.
confusion matrix
A set of charts that display total predictions generated for a dataset model and the percentage of true and false predictions by comparing predictions with actual results.
context information
In AIAMFG, context information is ERP data, such as the work order, operation, equipment instance, and so on, added to sensor data.
contextualization
The act of adding context information to sensor data.
CSV file
Comma-Separated Values file. CSV files can be opened by any spreadsheet program. Each row is a data record and the columns represent data fields.
data ingestion
To upload structured enterprise and semistructured sensor data into AIAMFG using comma separated values (CSV) file templates.
data lake
A location in the cloud that stores consolidated data.
data mining
The process of discovering patterns in large datasets.
data preparation
The process of creating a dataset that meets certain parameters.
dataset
A set of data that meets certain parameters, such as data related to a particular item and for a certain time period.
DCS (Distributed Control Systems)
A machine data acquisition system used to collect machine sensor data and provide the data in a CSV file format.
Decision Tree
An algorithm you can select when creating an insights model.
descriptive stats
The descriptive statistics for a model feature. For example, you can view descriptive statistics such as distinct count, mean, mode, and so on for a quality test result feature in a model.
entity
See also: business entity data.
ERP (Enterprise Resource Planning)
A type of software that organizations use to manage day-to-day business activities such as accounting, procurement, project management, risk management and compliance, and supply chain operations.
event
An occurrence that appears in the timeline viewer in relation to an entity. Includes Enterprise Resource Planning (ERP) events related to work orders, purchase orders, sales orders and other logistical movements of entities, and events captured by sensor devices.
exception
Something different than planned, such as a change in the planned work order completion date.
feature
A data attribute that you select to use as an input feature in a model.
flattened record
A denormalized record. The flattened record contains data attributes that were once stored in multiple tables. In AIAMFG, each record contains one work order and its related entities, such as item, operation, routing, and so on.
genealogy
The relationships between database objects. For example, for a particular product, view the related equipment used, sales orders, and other details.
histogram
A graph where each bar groups numbers into ranges. Taller bars show that more data falls in that range. A histogram is an approximate representation of the distribution of numerical data.
identity domain
Alogical namespace for users and groups, typically representing a discrete set of users and groups in the physical data store. Identity domains are used to identify the users associated with particular partitions.
insights
Insights from historical data analysis enable business users to discover the hidden patterns between influencing factors and production metrics such as production quality and yield
instance (equipment or resource)
A particular piece of equipment (or resource) when multiple pieces of the same equipment exists. For example, a filling machine, when multiple filling machines that are all exactly the same exist within the same plant.
instance (software application)
A bare metal or virtual machine (VM) compute host. The image used to launch the instance determines its operating system and other software. The shape specified during the launch process determines the number of CPUs and memory allocated to the instance.
key performance indicator
A quantifiable measure used to evaluate the success of an insights or predictions model..
Laboratory Information Management System (LIMS)
See also: LIMS
LIMS (Laboratory Information Management System)
Software that enables you to manage samples, test results and associated data to improve lab productivity.
machine learning
Computer algorithms that improve automatically through experience. Machine learning algorithms build a mathematical model based on sample data, known as "training data", in order to make predictions or decisions.
See also: artificial intelligence.
machine alert
A machine notification of an event such as downtime or the need for maintenance.
machine event detection
Also known as machine event analysis. Use imported machine sensor device data to identify machine events, such as actual start time.
MES (manufacturing execution system)
Computerized systems used in manufacturing to track and document the transformation of raw materials to finished goods. MES works in real time to enable the control of multiple elements of the production process
metadata
Data that provides information about other data. For example, an attribute's metadata could include the attribute's unit of measure.
metrics
Production measurements. For example, quality and yield measurements.
model
A mathematical model performed on a computer. A model is the algorithms and equations used to capture the behavior of the system being modeled.
module
A part of a program. Programs are composed of one or more independently developed modules.
network (supply chain network)
Visual display of the relationships between an object such as a purchase order, lot, serial number, work order, or sales order and the associated entities.
node
An object in a network.
operation
A step in a manufacturing process where you perform work on, add value to, and consume department resources for an assembly.
PAA (Piecewise Aggregate Approximation)
An approximation of a time series. Divide the original time-series into M equally sized frames and then compute the mean values for each frame. The sequence assembled from the mean values is the PAA approximation of the original time-series.
parametric data
Sample data from a population that can be adequately modeled by a probability distribution that has a fixed set of parameters.
prediction
The bin into which the target measure for a particular work order is most likely to fall, according to a particular predictions model.
prediction alert
A warning received if a product's target measure prediction falls within a bin that has alerts turned on.
predictor
An input variable or feature that influences a prediction.
Random Forest
An algorithm you can select when creating a Feature Significance model.
REST
REpresentational State Transfer. REST is a standard architectural style used when creating a web service.
REST API
Also known as a RESTful API. A REST API is a type of REST web service. Web services enable two machines to communicate over a network.
REST web service
A REST web service consists of the following components:
Persona assigned to an application user. Each role provides access to certain application modules, features and data relevant to their role.
rules (algorithm)
The definitions for a sequence of operations in an algorithm
SAX (Symbolic Aggregate approXimation)
A symbolic representation for complex time series data.
SCADA ( Supervisory Control and Data Acquisition)
Acomputer system for gathering and analyzing real time data. SCADA systems are used to monitor and control a plant or equipment.
seeded data
An initial set of data provided with a database when it is installed.
semi-structured data
A form of structured data that does not obey the formal structure of data models associated with relational databases or other forms of data tables, but nonetheless contains tags or other markers to separate semantic elements and enforce hierarchies of records and fields within the data. Examples of semi-structured data in AIAMFG are stream and alert data.
serial analysis
Models predict target measure results for serial units rather than for work orders.
serialized start operation
The operation where the serial number is applied to the item.
stratified sampling
Samples drawn from a number of separate strata of the population, rather than at random from the whole population, in order to provide a representative sampling.
stream data
Data that is continuously generated by different sources.
structured data
Organized elements of data that related to one another provided in a standardized format, such as a relational database.
SVM (Support Vector Machine)
An algorithm you can select when creating a Predictions model.
target measure
A production metric.
test data
Data used to assess the performance of a model. A model's predictions are based on test data.
time segment
A period of time. For example, 5 minutes.
time series data
An ordered data set with data points in specified intervals.
time series feature set
Used for model building and analysis. Constructed from time series data divided into time segments and either simple (average, min, max) or advanced functions (SAX).
timeline
Displays a measured value as it changes over time.
training data
Data used to fit the model. After receiving the model result based on the training data, adjust the model's parameters to best achieve the target measure.
unstructured data
Data that is not organized in a pre-defined manner or does not have a pre-defined data model.
web service
a service offered by an electronic device to another electronic device, communicating with each other over the internet.
work order
a product, service, task or a job that can be scheduled or assigned to someone.
Z-normalization
A calculation that measures the number of standard deviations by which the value of a data point is above or below the mean value of what is being observed or measured. Used with SAX.