80-20 Inventory Rule: A principle based on the Pareto Principle, stating that 80% of profits should come from 20% of inventory. Also known as the 80/20 rule, it highlights that 80% of effects result from 20% of causes.
Asset Control:
The process of managing long-term assets by organizing, tracking, auditing, and
recording essential
details. It involves adopting an effective asset management system, such as
asset management
software.
Asset Management Software:
A system storing inventory data, including quantity, location, value, and
condition. Also known as
asset tracking software, it helps streamline operations, identify inventory
loss, and centralize
asset information.
Asset Tagging:
An inventory management strategy involving affixing labels or "tags" to assets
to identify and
display additional information about each item.
Asset Tracking:
Continual management of business-owned assets, such as equipment and technology.
It helps monitor
assets as they change hands, locations, and depreciate over time.
ABC Stratification:
Ranking items based on a system (A-B-C) related to a characteristic, often used
by times sold.
Action Message:
Notifications in MRP, DRP, and planning systems indicating that an action needs
to be taken, such as
releasing or rescheduling an order.
Adaptive Smoothing:
A variation on exponential smoothing that automatically adjusts the smoothing
factor based on
forecast error.
Advanced Planning and Scheduling (APS):
A step beyond MRPII, it includes capabilities for finite capacity scheduling and
reacting to rapidly
changing demand.
Advanced Shipment Notification
(ASN):
A document notifying a customer of a shipment, often including PO numbers, SKU
numbers, lot numbers,
quantities, and container details.
Allocations:
Actual demand created by sales or production orders against a specific item.
Anticipation Inventory:
Inventory buildup to meet demand during a period of time when demand exceeds
capacity.
APICS:
The Association for Operations Management, formerly known as the American
Production and Inventory
Control Society.
Average Forecast Bias:
The average of a series of forecast errors.
Bill of Materials (BOM):
Itemized list of raw materials, parts, and components needed for product
manufacturing. It includes
quantities and secondary requirements like manuals and packaging.
Backflush:
Method for automatically issuing materials to a production order. Material is
issued when production
is posted against an operation. Quantities are calculated through the BOM,
reducing on-hand
balances.
Backorder:
Quantity on an order not filled on the required date due to inadequate inventory
levels.
Best of Breed:
Term used for highly functional, independent software modules from different
vendors. It recognizes
that a single software suite may not have the best modules for every business
area.
Blockchain:
Distributed database for secure, transparent, and tamper-proof data sharing. It
records transactions
in a way resistant to modification, enhancing transparency and traceability.
It's useful for
tracking items in the supply chain.
Black-box Forecasting:
Automated forecasting system making decisions without human input.
Bullwhip Effect:
Phenomenon amplifying demand variation up the supply chain. It was triggered by
an increase in
demand at a low level in the supply chain.
Consignment Inventory:
Inventory in the possession of the customer but owned by the supplier. The
customer purchases after
reselling or consuming.
Capacity:
The capabilities of a process, machine, location, or facility.
Capacity Requirements Planning
(CRP):
A detailed capacity planning tool that verifies resource ability to meet
scheduled production.
Carrying Costs:
Costs associated with specific inventory quantities, including investment and
storage costs.
Cheats:
Mathematical calculations for desired results, not necessarily precise or
correct, used in inventory
management.
Cost of Goods Sold(COGS):
Total cost of products sold during a specific period, expensed when products are
sold.
Collaborative Planning:
Planning strategy where trading partners share inventory planning data.
Component:
Any item used to produce another item.
Composite Forecast:
A forecast created by combining results of multiple forecasting methods.
Configuration Processing:
Software functionality allowing product definition by selecting predefined
options.
Configure-to-Order:
Manufacturing strategy between make-to-stock and make-to-order, stocking
components in anticipation
of customer orders.
Cost of Capital:
Costs associated with having money tied up in inventory, typically the interest
rate on business
debt.
Cumulative Forecast Bias:
The sum of a series of forecast errors.
Cumulative Lead Time:
The longest sum of consecutive lead times.
Current-Demand Inventory:
Inventory carried to meet immediate expected demand.
Cycle Count:
Process verifying inventory correctness by counting portions regularly, not the
entire inventory at
once.
Demand:
The need for a specific item in a specific quantity.
Demand Override:
Adjustment to supersede demand history, used for forecasting or safety stock
calculation. It can be
a fixed quantity or a factor.
Demand Planning Software:
Software with forecasting capabilities for determining order quantities and
safety stock.
Demand Variability:
Changes in demand due to trend, seasonality, events, and noise.
Dependent Demand:
Demand created as a direct result of another item's demand or demand from
another facility.
Deseasonalized:
Remove seasonality effect on demand.
De-trend:
Remove trend effect on demand.
Direct Shipping (Drop-shipping):
Procurement strategy allowing product sale without stocking; shipped directly
from supplier to
customer.
Distribution Inventory:
Result of distribution network, increased inventory due to multiple distribution
points.
Distribution Requirements Planning
(DRP):
Process for determining inventory requirements in a multiple facility
environment. Works with both
distribution and manufacturing.
Dock-to-Stock Cycle Measurement:
Measure of time between arrival at the dock and availability for sale or use.
Double Exponential Smoothing:
Forecasting method using exponential smoothing for both demand and trend.
DRP Relationship:
Also called bill of distribution, specifies item or facility-based supply
relationship between
facilities, used by DRP to flow demand.
Economic Order Quantity (EOQ):
Determines the optimal order quantity to meet demand while minimizing holding
and ordering costs.
Eaches:
Unit of measure where each individual piece is tracked as a quantity of one in
the computer system.
Effective Dates:
Dates on bills of materials and routings indicating when they should be included
in planning and
execution activities.
Effective Lead Time:
Lead time adjusted to consider additional factors, such as time between ordering
opportunities.
Engineer-to-Order:
Manufacturing strategy for custom products, involving design or engineering
tasks as part of the
order process.
Enterprise Resource Planning
(ERP):
Software systems managing various aspects of manufacturing or distribution
enterprises.
Event Index:
Number describing the relationship of demand over the time affected by an event,
e.g., promotion.
Excess Inventory:
Inventory greater than the "right amount" needed.
Exponential Smoothing:
Forecasting method using a weighted moving average and a smoothing factor.
Exponentially Smoothed Absolute
Deviation (ESAD):
Variation of Mean Absolute Deviation (MAD) using exponential smoothing on
absolute values.
Exponentially Smoothed Forecast
Bias:
Application of exponential smoothing to a series of forecast errors.
Fair Share Distribution:
Method of dividing inventory at a supplying facility among facilities based on
demand.
First-in, First-out (FIFO):
An inventory valuation method where the first goods purchased are the first to
be sold. This ensures
that older inventory is shipped out before newer inventory, providing a more
accurate estimation of
prices or values for each piece of inventory.
Fill Rate:
Success rate in filling orders, stated as line fill, order fill, unit fill, or
dollar fill.
Finished Goods:
Inventory in a salable or shippable form based on its location within the supply
chain. May vary
between supplying and receiving plants.
Finite Capacity Scheduling:
Manufacturing planning system scheduling within the capacity constraints of work
centers.
Firm Planned Order:
Special status preventing MRP from replanning a planned order.
Fixed Reorder Point:
Also called fixed order point, a preset quantity triggering the need for a new
order.
Fixed-Schedule Ordering System:
Replenishment system where orders are placed based on a predetermined schedule,
e.g., ordering every
Tuesday for a Friday delivery.
Flattening:
Removing levels from bills of materials, making the first item as part of the
manufacturing process
for the second item in the same production order.
Forecast:
Estimation of future demand.
Forecast Basis:
Data and information used to produce the forecast, often historical demand.
Forecast Bias:
Tendency of a forecast to be high or low.
Forecast Consumption:
Depleting the forecast as actual orders are received.
Forecast Error:
Measurement representing forecast error/accuracy, commonly calculated as
([Forecasted Sales] -
[Actual Sales]) / [Actual Sales].
Forecast Horizon:
Length of time into the future over which the entire forecast is based,
considering cumulative lead
time and other planning factors.
Forecast Interval:
Length of time over which each forecast period is based, following standard
cumulative measures of
time.
Forecast Override:
Adjustment used to supersede the normal forecast, either a fixed quantity or a
factor adjusting the
forecast.
Forecast Period:
Specific span of time described by a forecast quantity, often referred to as
"time buckets."
Freight Terms:
Agreement between a supplier and customer describing responsibility for
transportation costs.
Generic Software:
Products designed for broad environments. It focuses on standard business
practices for general
industry.
GMROII (Gross Margin Return on Inventory
Investment):
Calculation showing margin relative to average inventory investment. It's
calculated by dividing
annual gross margin by average inventory investment. Also, applicable to
individual items or groups.
Gross Margin:
Difference between cost and sell price.
Gross Requirements:
Total demand (dependent and independent) for an item within a specific time
period. It's used to
calculate net requirements.
Group Logic:
Methods to manage inventory based on groups of items rather than single items.
Hedge Inventory: Inventory purchased to anticipate events making necessary items challenging or expensive. Used to protect against or capitalize on price fluctuations due to factors like seasonal variations, supply and demand imbalances, exchange rate changes, or special promotions.
Inventory:
Any quantifiable item that a business handles, buys, sells, stores, consumes,
produces, or tracks.
Inventory Asset:
An owned and reusable item such as equipment, tools, machinery, vehicles, etc.,
used for day-to-day
operations.
Inventory Automation:
The use of inventory management software and technology, including barcodes or
QR codes, to track,
analyze, and control items efficiently.
Inventory Cycle Counting:
A strategy where defined portions of inventory are counted on a rotating
schedule, spreading the
count throughout the year to avoid the need for a single comprehensive count.
Inventory List:
A complete, itemized list of every product in stock, including raw materials,
work-in-progress, and
finished goods.
Inventory Management:
The process of ordering, organizing, storing, and utilizing a business’s
inventory to keep
operations running smoothly.
Inventory Optimization:
Maintaining an optimal inventory level to meet customer demand, reducing costs,
and avoiding issues
like stockouts and excess inventory.
Inventory Turnover:
The measure of how many times a company has sold and replaced inventory within a
specific period,
indicating efficiency.
IF-THEN-ELSE:
The common logic used by software to make decisions, describing actions based on
whether a given
situation is true or false.
IF Statement:
A calculation using IF-THEN-ELSE logic, often referring to a spreadsheet (Excel)
formula or similar
program.
Independent Demand:
Demand not created as a result of another item’s demand or demand from another
facility.
Industry-specific Software:
Software designed for a specific industry, focusing on typical business
practices of that industry.
Infinite Capacity Scheduling:
A manufacturing planning system ignoring capacity constraints and scheduling
based purely on demand.
In-transit Quantity:
Quantity shipped from one facility but not yet received into another.
Inventory Characteristic:
Any trait describing the types of inventory, such as physical size, form, demand
patterns, and
costs.
Inventory Classification:
Logical grouping of inventory based on user-defined characteristics.
Inventory Management:
Control of inventory to best achieve business objectives, involving both
physical management and
data/systems management.
Inventory System:
Collection of programs and data used to plan and track inventory balances and
activities.
Inventory Turns:
A measure of inventory velocity calculated by dividing average/current inventory
investment into
annual cost of goods sold (COGS).
ISM (Institute for Supply
Management):
Formerly known as National Association of Purchasing Management (NAPM), it is an
organization for
supply management professionals.
Issue:
To reduce on-hand inventory and assign it to a specific document or process,
such as issuing raw
materials to a production order or finished goods to a sales order.
Item:
Any unique configuration of a material or product managed as part of inventory,
used synonymously
with SKU.
Item Master:
A collection of data describing a specific item, also used to describe the
database table containing
this data.
Item Number:
The identification number assigned to an item, also called part number, SKU
number, or SKU.
Item Numbering Scheme:
The format or template used for assigning item numbers.
Just in Time Inventory (JIT): An inventory management system focused on minimizing inventory levels to reduce costs, enhance efficiency, and minimize waste. JIT, or just-in-time, is a strategy optimizing manufacturing processes by eliminating process waste, such as wasted steps, materials, and excess inventory. It is often synonymous with "Lean manufacturing" or the "Toyota production system."
Kitting:
Inventory management strategy bundling individual items into a kit. Tracked,
used, and sold as a
single piece despite multiple parts.
Kanban:
Replenishment system triggered by emptying a container. Physical notification
(card or empty
container) sent for refilling to the previous operation or supplier.
Kit:
Items made up of multiple separate parts, not assembled.
Landed Cost:
The total cost of shipping a product, including taxes, fees, customs, risk, and
overhead. Also
referred to as "true cost" or "delivered cost."
Lead Time:
The duration it takes to receive an order after placement. Lead-time demand is
the expected demand
during this period.
Last Mile:
The final stage of the supply chain, involving delivery to the customer's
doorstep. It is often the
most expensive and time-consuming part of the shipping process.
Last-Period Demand:
A forecasting method using demand from the previous period as a forecast for
subsequent periods.
Last-Relative-Period Demand:
A forecasting method using the relative period (usually from the previous year)
to forecast demand.
Law of Large Numbers:
Observation that larger numbers generally result in lower variability.
Lead-Time Factor:
A multiplier adjusting standard deviation based on forecast periods to estimate
standard deviation
based on lead time.
Lead-Time Offsetting:
The process of offsetting dependent demand items based on the lead time of
parent items or
facilities.
Lean:
Also known as Just-In-Time (JIT), emphasizing efficiency by minimizing waste and
maintaining low
inventory.
Legacy System:
Outdated computer system, often custom-built or modified over the years.
Level:
Also known as normalized demand or base demand, it represents the starting point
for a forecast.
Last-in First-out (LIFO):
An inventory valuation method where the latest purchased or produced goods are
the first to be
expensed.
Line Item:
A single detail record, commonly used in sales orders or purchase orders.
Lot-for-Lot:
Basic lot sizing method using demand during the planning time period as the lot
size.
Lot Size:
The quantity of an item ordered for delivery on a specific date or manufactured
in a single
production run.
Lot-Size Inventory:
Result of ordering or manufacturing more inventory than required to meet current
demand and safety
stock.
Minimum Order Quantity (MOQ):
The smallest amount of product a supplier sells to a business placing an order.
Moving Average Cost:
The cost of existing inventory on hand plus the cost of new inventory ordered
divided by the exact
number of items in stock.
MRO Inventory (Maintenance, Repair, and
Operations
Inventory):
Inventory utilized by a business for preventive and corrective maintenance on
assets or to keep
day-to-day business activities running efficiently.
MAD (Mean Absolute Deviation):
Average of the absolute values of a series of variances, used in forecast error
measurement and
safety stock calculations.
Make-to-Order:
Manufacturing strategy where products are not manufactured until actual orders
are received,
reducing the need for finished goods inventory.
Make-to-Stock:
Manufacturing strategy where adequate finished goods inventory is carried to
meet forecasted demand.
Management by Exception:
A strategy automating most decisions and monitoring for exceptions.
Manufacturing Execution System
(MES):
Software integrating with enterprise systems to enhance shop floor control
functionality.
Manufacturing Lead Time:
Combination of setup time, run time, move time, and queue time in the
manufacturing process.
Master Production Schedule (MPS):
Planning tool balancing demand with capacity by moving production to periods
with available
capacity.
Modification:
A change to software requiring changing or adding to the software code.
Move Time:
Time taken to physically relocate materials from one manufacturing operation to
the next.
Multi-bin System:
Replenishment system using two or more physical bins for each item.
Multi-level Bill of Materials:
A bill-of-materials structure where components have their own bills below them,
logically linked for
planning purposes.
Multi-Period Forecast Error Amplitude
Measurement
(MPFEAM):
Measurements quantifying the size of forecast errors over multiple forecast
periods.
Multi-Plant:
Environments where multiple facilities are managed.
Multi-Plant MRP:
MRP extended with DRP logic to plan inventory in multi-plant environments.
Nervousness (in MRP systems):
Frequent changes to planned orders for lower-level items resulting from minor
changes in demand for
higher-level items.
Net Change MRP:
A process recalculating gross and net requirements and planned orders for items
with changes in
planning data. Can be run as a batch or real-time in a live environment.
Net Requirements:
The result of adjusting gross requirements by current on-hand, safety stock, and
inbound quantities.
Netting:
The process of adjusting gross requirements by current on-hand, safety stock,
and inbound
quantities.
Noise (in demand):
Unpredictable variation in demand not attributed to trend, seasonality, or other
predictable
factors.
Non-stock Inventory:
Inventory not tracked in the perpetual inventory system. May lack an item-master
record or internal
SKU number. Can also refer to order-as-needed inventory with an item-master
record and SKU, but no
carried stock.
Normal Distribution (in statistical
analysis):
A distribution model with a bell-shaped curve representing the probability of an
occurrence.
Commonly used to predict demand variability based on historical data.
Normalize:
To remove elements like seasonality, trend, or event effects from demand.
Obsolescence:
The process when inventory becomes obsolete.
Obsolete:
The condition of being no longer useful due to outdated designs or the passage
of time.
On-time Delivery:
A fill-rate measurement indicating the percentage of orders filled by the
promised date.
Operation:
Combination of a physical facility and its processes. Also, referred to as a
step in the
manufacturing process.
Optimization:
The process of obtaining the best practical result from a given problem.
Optional Reorder Point:
A higher reorder point used to avoid downtime or meet specific requirements.
Order as Needed:
A replenishment method triggering orders only when actual demand is present.
Order Cost:
Sum of fixed costs incurred with each item order or production.
Order Cycle:
The time between receipts of an item or the duration an ordered quantity should
last.
Ordering System:
An inventory ordering system ensures efficient stock replenishment by
determining optimal reorder
points and quantities to meet customer demand while minimizing holding costs.
Examples are Fixed
reorder point, min-max, multi-bin, MRP, DRP.
Outsourcing:
Transferring responsibilities for a process to an outside supplier.
Overhead:
Indirect costs associated with facilities and management applied to
manufacturing costs.
Override:
An adjustment used to supersede standard results in decision-making or
calculations.
Pick List:
Document specifying inventory for order fulfillment.
Purchase Requisition:
Internal document informing the purchasing department of intended purchases.
Pareto Principle (80/20 Rule):
Small causes responsible for a majority of effects.
Part Number (Item Number):
Identifier for components in production.
Parts List (Materials List):
Inventory of materials needed for production.
Payment Terms:
Agreement specifying payment details between supplier and customer.
Period Order Quantity (POQ):
Ordering method based on time rather than units.
Periodic Review Inventory System:
System reviewing and replenishing inventory periodically.
Perpetual Inventory System:
System adjusting on-hand balances through transactions.
Phantom Bill of Materials:
Fictitious BOM for common subassemblies or kits.
Physical Inventory:
Counting all inventory in a single event.
Planned Order:
System-generated order recommendation in MRP and DRP.
Planning Bill of Materials (Super
Bill):
Fictitious BOM grouping products or options.
PO (Purchase Order):
Document authorizing, tracking, and processing purchased items.
Postponement:
Delaying specific operations until just before shipping.
Predictor Variable:
Data set predicting another set in regression analysis.
Procure-to-Order:
Inventory strategy procuring products after receiving customer orders.
Procure-to-Stock:
Inventory strategy maintaining finished goods inventory for forecasted demand.
Product Life Cycle:
Period an item is considered actively saleable.
Product Life Cycle Index:
Number describing demand over the complete product life cycle.
Production Order:
Document for processing a production run.
Production Plan:
High-level, long-term plan for what will be produced.
Production Run:
Physical execution of tasks for a production order.
Promised Date:
Date a supplier expects to fulfill a customer order.
Pull System:
Ordering or production system driven by actual customer demand.
Purchase Order (PO):
Document authorizing, tracking, and processing purchased items.
Purchasing Contract:
Legal document on pricing and terms for significant sales transactions.
Push System:
Ordering or production system triggered by expected demand.
Pick List:
Document specifying inventory for order fulfillment.
Purchase Requisition:
Internal document informing the purchasing department of intended purchases.
Pareto Principle (80/20 Rule):
Small causes responsible for a majority of effects.
Part Number (Item Number):
Identifier for components in production.
Parts List (Materials List):
Inventory of materials needed for production.
Payment Terms:
Agreement specifying payment details between supplier and customer.
Period Order Quantity (POQ):
Ordering method based on time rather than units.
Periodic Review Inventory System:
System reviewing and replenishing inventory periodically.
Perpetual Inventory System:
System adjusting on-hand balances through transactions.
Phantom Bill of Materials:
Fictitious BOM for common subassemblies or kits.
Physical Inventory:
Counting all inventory in a single event.
Planned Order:
System-generated order recommendation in MRP and DRP.
Planning Bill of Materials (Super
Bill):
Fictitious BOM grouping products or options.
PO (Purchase Order):
Document authorizing, tracking, and processing purchased items.
Postponement:
Delaying specific operations until just before shipping.
Predictor Variable:
Data set predicting another set in regression analysis.
Procure-to-Order:
Inventory strategy procuring products after receiving customer orders.
Procure-to-Stock:
Inventory strategy maintaining finished goods inventory for forecasted demand.
Product Life Cycle:
Period an item is considered actively saleable.
Product Life Cycle Index:
Number describing demand over the complete product life cycle.
Production Order:
Document for processing a production run.
Production Plan:
High-level, long-term plan for what will be produced.
Production Run:
Physical execution of tasks for a production order.
Promised Date:
Date a supplier expects to fulfill a customer order.
Pull System:
Ordering or production system driven by actual customer demand.
Purchase Order (PO):
Document authorizing, tracking, and processing purchased items.
Purchasing Contract:
Legal document on pricing and terms for significant sales transactions.
Push System:
Ordering or production system triggered by expected demand.
Raw Materials (Components):
Inventory used in the manufacturing process. It includes bulk materials like
ore, paper, plastic, or
steel.
Receipts:
Materials or transactions associated with the receiving process.
Receiving:
Process of placing materials into inventory. It describes the department where
receiving activities
take place.
Regenerative MRP:
Process of completely regenerating planned orders for all items. It wipes out
existing planned
orders, recalculates gross and net requirements.
Regression Analysis:
Techniques to determine a mathematical relationship between sets of data. It
involves predicting one
set of data (predictor variable) based on others.
Relevant History:
Data recorded under business conditions similar to current/future conditions. It
used to effectively
forecast future demand.
Reorder Point:
Inventory level triggering an order for a specific item. It's calculated as
expected usage during
lead time plus safety stock.
Replenishment:
Process of moving inventory within a warehouse or plant. It involves moving
inventory between
facilities or from suppliers to meet demand.
Requested Date:
Date a purchased item is requested to be received at a customer's location.
Requisitions:
Documents created by individuals without purchasing authority. It specifies
item, quantity, cost,
terms, and vendor information.
Resources Requirements Planning
(RRP):
Capacity planning tool verifying key resource ability for production/business
plans. It focuses on
long-term planning for facilities, major equipment, capital, and workforce.
Restock:
An act of replenishing inventory by ordering additional stock to meet customer
demand and maintain
optimal supply levels.
Rough-Cut Capacity:
Capacity planning tool verifying key resource ability to meet master production
schedule. It's
intermediate level between Resources Requirements Planning (RRP) and Capacity
Requirements Planning
(CRP).
Routing:
List of operations/steps to complete a manufacturing process. It's used with the
bill of materials
to specify steps, including labor and machine requirements.
Run Time:
Time taken to produce a single unit in an operation step. It excludes setup time
or queue time. It
can accumulate for multiple operations or units in a production order.
Safety Stock:
Extra inventory held in anticipation of unexpected disruptions or delays.
Supplier Relationship Management
(SRM):
Strategic approach to optimizing and coordinating business relationships with
suppliers for maximum
profitability.
Safety Lead Time:
Additional time incorporated into lead time calculations for ordering, excluding
from the final
requested date calculation.
Sales Order:
Document approving, tracking, and processing outbound customer shipments.
Supply Chain Management(SCM):
SCM focuses on optimizing processes for efficient production, distribution, and
delivery of goods or
services.
Scrap:
Inventory to be discarded or recycled due to manufacturing processes or damage
during storage.
Scrap Rate:
Expected rate of scrap for specific components in the manufacturing of an item.
Seasonality:
Fluctuations in demand with a repeating pattern over equivalent time periods.
Seasonality Index:
Number indicating the relationship of each period's demand to the average demand
over a complete
seasonal cycle.
Self-Induced Seasonality:
Repeating demand patterns caused by internal processes and policies.
Semi-Processed Materials:
Stockable items that undergo partial processing before additional processing.
Service Factor:
Multiplier in statistical safety stock calculations associated with a desired
service level.
Service Level:
Key input to statistical-based safety stock calculations and a measure of
fill-rate and on-time
delivery.
Setup Costs:
Costs associated with initiating a production run, including labor, machine
time, and setup-related
expenses.
Setup Time:
Time required to prepare equipment and materials for a production run.
Shipping Order:
Document approving, tracking, and processing outbound shipments.
SKU (Stock-Keeping Unit):
Specific item in a specific unit of measure, identified by a unique number.
Smoothing:
Removing variation from demand.
Smoothing Factor:
Number (0.01 to 0.99) used to weight recent demand against forecast in
exponential smoothing
calculations.
Sourcing:
Activity of finding suppliers for products, materials, or services.
Spreadsheet Modeling:
Creating a mathematical representation of a business problem in a spreadsheet.
Square Root Trick:
Statistical tool softening a ratio by taking the square root.
Standard Business Practices:
Typical practices for general or specific industries.
Standard Deviation:
Statistical term describing the spread of variation in a distribution.
Stocking Type:
Classification in planning and execution systems identifying primary stocking
characteristics.
Stocking Unit of Measure:
Unit of measure used to track inventory within a facility.
Stockout:
Situation with inadequate inventory levels to meet current demand.
Storage Cost:
Costs associated with the physical storage of inventory, including space and
equipment.
Subassembly:
Stockable item that undergoes assembly and is used in the assembly of other
items.
Supply-Chain Optimization
Software:
Software with advanced optimization algorithms for inventory, transportation,
and multi-facility
planning in complex supply chains.
Terms:
Agreement between a supplier and customer outlining sale conditions, including
payment, freight,
change-of-ownership, and return policies.
Time Fence:
Time frame for controlling changes to production schedules, usually implemented
as guidelines.
Time-phased Order Point:
System using immediate forecast and actual customer orders to trigger reorder
decisions.
Total Quality Management (TQM):
Management strategy focusing on continuous improvement.
Tracking Signal:
Calculation describing the overall health of the forecast relative to trend,
used to initiate
changes in forecasting technique or parameters.
Transportation Inventory:
Inventory currently in-transit, from shipper's facility to consignee's facility.
Trend:
Gradual increase or decrease in demand over time.
Trend Adjustment:
Mathematical calculation to adjust future forecasts for trend extension.
Trend Element:
Specific aspect of trend, e.g., changes in market share.
Trend Lag:
Forecast's inability to adequately account for trend, often due to smoothing
calculations or lack of
proper trend extension.
Triple Exponential Smoothing:
Forecasting method using exponential smoothing for demand, trend, and
seasonality index.
Two-bin System:
Simplistic replenishment system using two physical bins for each item, sent for
refill when one is
emptied.
UOM Inventory:
Unit-of-Measurement inventory is standardized, physical units for quantifying
stock in a business.
Unfinished Goods:
Items used to produce finished goods, known as components, ingredients, raw
materials,
semi-processed materials, and subassemblies.
Units:
Individual pieces of physical inventory that constitute the quantity of an item.
Unit of Measure:
Describes how the quantity of an item is tracked in the inventory system (e.g.,
eaches, cases,
pallets, pounds).
Unit-of-Measure Conversion:
Conversion ratio for handling multiple units-of-measure with the same item.
Vendor Managed Inventory (VMI):
Process where a supplier manages customers' inventory levels and purchases of
supplied materials.
Vendor Managed Inventory (VMI): A supply chain arrangement where a qualified third-party, such as a supplier or manufacturer, controls inventory and related decisions on behalf of the seller. VMI is often used as an abbreviation for Vendor Managed Inventory.
Warehouse Management:
Refers to daily operations ensuring smooth warehouse functioning, including
inventory ordering,
tracking, and management during movement, usage, or sale.
Warehouse Management System (WMS):
Software providing visibility into inventory, assets, logistics, and fulfillment
operations.
Weighted Moving Average:
Forecasting method where different weights (totaling 1) are applied to
historical periods to
calculate the forecast.
Work-in-Process (WIP):
Financial account representing the value of inventory, labor, and overhead
issued to production but
not yet resulting in a finished product.
Work Order (or Production Order):
Instruction for manufacturing or production processes.
Yield Rate: The anticipated success rate in the manufacturing process for an item. It is measured at the parent-item level, contrasting with the scrap rate, which is assessed at the component level.
80-20 Inventory Rule: A principle based on the Pareto Principle, stating that 80% of profits should come from 20% of inventory. Also known as the 80/20 rule, it highlights that 80% of effects result from 20% of causes.
Active Learning:
A machine learning approach where the model selects the most informative data to
learn from.
AGI (Artificial General
Intelligence):
Advanced AI capable of outperforming humans in tasks and self-improving.
AI (Artificial Intelligence):
Technology that simulates human intelligence to perform tasks.
AI Augmentation:
Using AI to enhance, rather than replace, human decision-making and tasks.
AI Bias Mitigation:
Techniques to reduce or eliminate bias in AI outputs.
AI Ethics:
Guidelines ensuring AI operates without harming humans or exacerbating biases.
AI Fairness:
Ensuring that AI systems do not reinforce or create discrimination.
AI Governance:
Policies and frameworks to ensure responsible AI development and use.
AI Model Robustness:
Ensuring an AI system performs reliably across a variety of conditions.
AI Orchestration: Coordinating
multiple AI models to work together on complex tasks.
AI Safety:
A field focused on preventing AI from evolving into a potential threat to humanity.
Algorithm:
A set of instructions used by AI to recognize patterns and learn from data.
Alignment:
Adjusting AI to ensure it generates desired outcomes.
Anthropomorphism:
Attributing human traits to AI, like thinking a chatbot is sentient.
Autonomous Agents:
AI systems capable of performing tasks independently, like self-driving cars.
Bias:
Systemic errors in AI models due to biased training data.
Black Box Problem:
The lack of transparency in AI decision-making processes.
Chatbot:
AI that simulates human conversation through text.
ChatGPT:
An AI chatbot developed by OpenAI using a large language model (GPT) to generate
human-like responses.
Claude:
An AI chatbot developed by Anthropic, focusing on safety and explainability in large
language model outputs.
Cognitive Computing:
Another term for AI, emphasizing mimicking human thought processes.
Convolutional Neural Network (CNN):
A neural network type specialized in image and video recognition.
Data Augmentation:
Enhancing AI training by remixing or adding data.
Data Governance:
Policies ensuring AI training data is managed ethically and responsibly.
Data Labeling:
The process of categorizing data to help AI learn patterns.
Data Privacy:
Protecting personal data from misuse by AI systems.
Data Wrangling:
Cleaning and structuring raw data into a usable format for AI.
Deep Learning:
A type of machine learning using neural networks to recognize complex patterns.
Diffusion Models:
AI techniques that add random noise to data and then remove it for image generation.
Digital Twin:
A virtual replica of a physical object or system used in AI simulations.
Edge AI:
AI models running on local devices instead of centralized servers for faster
decision-making.
Embedding:
A method where data is represented as continuous vectors in machine learning.
Emergent Behavior:
AI exhibiting unintended abilities or behaviors.
End-to-End Learning (E2E):
Training models to solve tasks from start to finish without sequential steps.
Ethical Considerations:
Understanding the moral implications of AI usage, including privacy and fairness.
Exploratory Data Analysis (EDA):
Analyzing datasets to summarize their main characteristics.
Explainability:
The degree to which AI decisions can be understood and explained.
Few-Shot Learning:
Training an AI with only a small amount of data, yet achieving accurate results.
Federated Learning:
A technique where AI models learn across decentralized data sources.
Foom:
A theory where AGI rapidly outpaces human control, potentially endangering humanity.
GANs (Generative Adversarial Networks):
AI models that create new content by pitting two neural networks against each other.
Generative AI:
AI that creates content like text, images, and code from data patterns.
Generative AI Art:
AI-generated visual art created through models like GANs.
Google Gemini:
Google’s AI chatbot, similar to ChatGPT but connected to live web data.
Gradient Descent:
A method used to optimize AI models by adjusting parameters.
Gradient Explosion:
The opposite of vanishing gradient, where gradients grow too large during training.
GPT (Generative Pretrained
Transformer):
A class of large language models like ChatGPT, trained on vast amounts of data to
generate human-like text.
Guardrails:
Policies to ensure AI behaves ethically and doesn’t produce harmful content.
Hallucination:
AI providing false or nonsensical answers confidently.
Hybrid AI:
Combining rule-based systems with learning-based AI for enhanced decision-making.
Hyperparameters:
External configurations that control an AI model’s learning process.
Inference: The process of using a trained AI model to make predictions or decisions.
Latent Space:
An abstract space where AI models learn to represent and manipulate data.
Large Language Model (LLM):
AI models trained on vast text data to generate human-like responses.
LLAMA (Large Language Model Meta AI):
Meta’s family of open-source large language models, designed for both research and
commercial use.
Machine Learning (ML):
A subset of AI where computers improve predictions by learning from data.
Meta-Learning:
AI learning how to learn, improving its ability to generalize across tasks.
Microsoft Bing AI:
AI-powered search engine using technology similar to ChatGPT.
Model Compression:
Techniques to reduce the size of AI models for efficiency.
Model Drift:
When an AI model’s performance deteriorates over time due to changes in input data.
Multimodal AI:
AI that can process multiple inputs like text, images, and speech.
Natural Language Processing (NLP):
AI's ability to understand and generate human language.
Natural Language Understanding (NLU):
AI’s ability to comprehend the meaning and intent behind language.
Neural Architecture Search (NAS):
A technique for automating the design of neural networks.
Neural Network:
AI structure mimicking the human brain to recognize patterns and learn.
Neuro-Symbolic AI:
Combining neural networks with symbolic reasoning for more robust AI
decision-making.
Overfitting: A model too closely aligned with training data, failing to generalize to new data.
Parameters:
Numerical values that shape an AI model’s decision-making process.
Paperclip Maximizer:
A thought experiment where AI’s single-minded goal leads to catastrophic
consequences.
Personalization:
Tailoring AI outputs to individual preferences and needs.
Pretraining:
The initial phase where an AI model learns basic patterns before fine-tuning on
specific tasks.
Prompt:
Input text given to an AI model to generate a response.
Prompt Chaining:
AI using previous interactions to improve or influence future responses.
Quantum AI: The integration of quantum computing with AI to solve complex problems faster.
RAG (Retrieval-Augmented Generation):
A model that enhances generative AI by retrieving relevant information from external
sources to improve responses.
Recurrent Neural Network (RNN):
A neural network where connections form cycles, useful for sequential data.
Recommender System:
AI algorithms suggesting products or content based on user behavior.
Reinforcement Learning:
A technique where AI learns by receiving feedback from its actions to improve over
time.
Self-Attention Mechanism:
A method used in transformers where every part of the input is evaluated relative to
the others.
Self-Supervised Learning:
An AI training method where the model generates its own labels from the data.
Simulated Annealing:
An optimization technique inspired by the process of heating and slowly cooling
material to reduce defects.
Sparse Data:
Datasets with missing or irregular entries, challenging for AI training.
Stochastic Parrot:
A metaphor for AI's ability to generate language without understanding its meaning.
Style Transfer:
AI’s ability to apply the visual style of one image to another.
Synthetic Data:
Artificially generated data used to train AI when real data is scarce.
Synthetic Media:
Media content, such as images or videos, generated by AI.
Temperature:
A setting that controls how unpredictable or creative an AI’s responses are.
Text-to-Image Generation:
AI creating images based on written descriptions.
Tokenization:
The process of breaking text into smaller pieces, or tokens, for AI to process.
Tokens:
The smallest units of text an AI processes to generate language responses.
Training Data:
Data used to teach AI models, including text, images, and code.
Transfer Learning:
When a model trained on one task is adapted for another, reducing training time.
Transferable Adversarial Attacks:
A security risk where attacks designed for one model work on others.
Transformer Model:
An AI architecture that understands context by analyzing relationships within data.
Turing Test:
A test to determine if AI responses are indistinguishable from human communication.
Unsupervised Learning: AI learning without labeled data, finding patterns on its own.
Vanishing Gradient Problem:
An issue in deep learning where gradients become too small for effective learning.
Vision AI:
AI specialized in interpreting visual data, such as images or video.
Weak AI (Narrow AI): AI focused on specific tasks without broader learning capabilities.
Zero-Shot Learning: AI completing a task without having been trained specifically for it.
51% Attack: An attack where a group of miners controls more than half of the network’s computational power, potentially allowing them to reverse transactions or double-spend tokens.
Account:
A key pair giving access to blockchain assets.
Address:
A public string for receiving blockchain assets.
Airdrop:
Token distribution for promotional purposes.
Altcoin:
Any cryptocurrency other than Bitcoin.
AMM (Automated Market Maker):
A DeFi protocol for trading without intermediaries.
Atomic Swap:
A trustless way to trade crypto across different blockchains.
Blockchain:
A decentralized ledger that records transactions in blocks.
Bridge:
Allows assets to move between different blockchain networks.
Bytecode:
A low-level language used by virtual machines like Ethereum's EVM.
Byzantine Fault Tolerance (BFT):
A blockchain's resilience to failures or attacks, ensuring consensus despite malicious actors.
BUIDL:
A term encouraging the active building of projects in the crypto space.
Cold Storage:
Offline storage of cryptocurrency to prevent hacking.
Consensus Mechanism:
Protocols like Proof of Work or Proof of Stake used to verify blockchain transactions.
Crypto:
Short for cryptocurrency, referring to digital currencies secured by cryptography.
Custodial Wallet:
A wallet controlled by a third-party, where the user does not own the private keys.
DAO (Decentralized Autonomous Organization):
Blockchain-based organizations governed by smart contracts.
dApp:
A decentralized application running on a blockchain without centralized control.
DeFi (Decentralized Finance):
A blockchain-based financial ecosystem operating without intermediaries.
Depin (Decentralized Physical Infrastructure Networks):
Networks that use blockchain for decentralizing hardware infrastructure.
DEX (Decentralized Exchange):
A peer-to-peer marketplace for cryptocurrencies without intermediaries.
Distributed Ledger:
A database spread across multiple nodes, ensuring decentralization.
ERC-20:
A technical standard for creating fungible tokens on Ethereum.
ERC-721:
A standard for creating non-fungible tokens (NFTs).
EVM (Ethereum Virtual Machine):
The engine that processes smart contracts on Ethereum.
Epoch:
A period during which a set of blocks is processed in Proof of Stake networks.
Fiat Currency:
Government-issued currency like USD or EUR.
Finality:
The point at which a transaction becomes irreversible on the blockchain.
Fork:
A blockchain split that creates two versions of the network.
Gas:
A fee for processing transactions on Ethereum, measured in Gwei.
Genesis Block:
The first block on a blockchain.
Governance:
The method by which blockchain stakeholders make decisions, often through voting using governance tokens.
Hard Fork:
A blockchain upgrade that is not backward-compatible.
Hash:
A unique string created by a hashing algorithm to secure blockchain data.
ICO (Initial Coin Offering):
A method of raising funds through the sale of tokens.
Immutable:
The characteristic of blockchain data being unchangeable once written.
IPFS (InterPlanetary File System):
A decentralized file storage protocol used for sharing and storing files securely on the blockchain.
JSON-RPC: A protocol used by Ethereum and other blockchains for communication between nodes and clients.
KYC (Know Your Customer): Regulations that require identity verification to prevent fraud.
Layer 2:
A secondary framework or protocol built on top of a blockchain to improve scalability and speed.
Liquidity Pool:
A pool of funds in a smart contract, allowing for decentralized trading and lending.
Mining:
The process of validating transactions and creating new blocks on Proof of Work networks.
Multi-chain:
A system where multiple blockchain networks interoperate, allowing asset transfers across chains.
NFT (Non-Fungible Token):
A digital asset representing ownership of a unique item or piece of content on a blockchain.
Node:
A computer that helps maintain the blockchain by validating and broadcasting transactions.
Off-chain:
Transactions or data that occur outside the blockchain, typically for efficiency.
On-chain:
Data and transactions that take place directly on the blockchain.
Private Key:
A cryptographic key used to sign transactions and access funds.
Public Key:
A cryptographic key used to receive funds and verify signatures.
Proof of Stake (PoS):
A consensus mechanism where validators are chosen based on the amount of cryptocurrency they hold.
Proof of Work (PoW):
A consensus mechanism requiring miners to solve computational puzzles to validate transactions.
Provenance:
The ability to trace the origin and history of an asset or transaction on the blockchain.
QR Code: A scannable code used in blockchain payments and transfers.
RWA (Real World Assets):
The tokenization of physical assets, such as real estate, on a blockchain.
Rollup:
A Layer 2 scaling solution that bundles transactions for more efficient processing on a blockchain.
Security Token:
A token representing ownership of an underlying financial asset, such as equity or bonds.
Seed Phrase:
A set of words that represents a private key, used to restore access to a wallet.
Sharding:
A scaling solution that splits blockchain data into smaller, more manageable parts.
Smart Contract:
Self-executing contracts coded onto a blockchain, which automatically enforce agreements.
Token:
A digital asset issued on a blockchain, representing anything from value to governance.
Token Gating:
Restricting access to services or content based on token ownership.
Tokenomics:
The study of the economic design and distribution of tokens within a blockchain ecosystem.
TPS (Transactions Per Second):
A measure of the throughput of a blockchain network.
Utility Token: A token used to access products or services within a blockchain ecosystem.
Validator:
A participant in a Proof of Stake network responsible for verifying transactions.
Volatility:
The degree of fluctuation in the price of a cryptocurrency.
Wallet:
A software or hardware tool used to store and manage private keys and interact with the blockchain.
Wallet:
A decentralized version of the internet built on blockchain technology.
xDai: A sidechain on Ethereum designed for low-cost, fast transactions.
Yield Farming: A DeFi activity where users earn returns by providing liquidity or staking tokens.
Zero-Knowledge Proof: A cryptographic method where one party can prove knowledge of a value without revealing the value itself.