Why These Terms Matter

Industrial and systems engineering sits at the intersection of manufacturing, quality, operations research, and management — which means its vocabulary is drawn from several traditions at once: Toyota Production System terms like kaizen and kanban, statistical quality control terms like Cp/Cpk, reliability engineering terms like MTBF, and formal operations-research terms like EOQ and linear programming. An industrial engineer moving between a plant floor, a Six Sigma project, and a capacity-planning spreadsheet needs fluency in all of them.

Precision matters because these terms drive real decisions — how much safety stock to carry, whether a process is capable of meeting a specification, how many operators a line needs, and what "world-class" performance actually looks like. This glossary defines 55 of the most frequently used terms across lean manufacturing, quality engineering, reliability, and industrial statistics, organized alphabetically with the standard or methodology each term comes from.

A

Andon — Toyota Production System
A visual signal — typically a light, board, or cord-pull — that alerts supervisors and support staff to a problem on the production line, such as a quality defect, parts shortage, or equipment malfunction. The classic "andon cord" gives any line worker the authority to stop the line when a defect is found, embodying the principle of jidoka (stopping at the first sign of a problem rather than passing defects downstream).
Automation, Fixed vs. Flexible — Manufacturing systems
Fixed (hard) automation performs a single, unchanging sequence of operations at high volume and low unit cost, but is expensive to reconfigure for a new product — think of a dedicated transfer line. Flexible (soft) automation, typically built around programmable robots and CNC machines, can be reprogrammed for different parts with modest changeover time, trading some throughput for adaptability. Most modern plants use a mix, chosen based on expected production volume and product variety (the classic volume-variety tradeoff).

B

Bottleneck — Theory of Constraints
The resource in a process — a machine, workstation, or step — whose capacity is lower than the capacity of every other resource, so it limits the throughput of the entire system. Eliyahu Goldratt's Theory of Constraints holds that improving any non-bottleneck resource does not increase overall output; only relieving the bottleneck (through the "Five Focusing Steps": identify, exploit, subordinate, elevate, repeat) increases system throughput.
Bill of Materials (BOM) — Manufacturing/ERP
A structured list of every raw material, component, subassembly, and quantity required to manufacture one unit of a finished product, often organized in levels (a single-level BOM lists only direct components; an indented/multilevel BOM shows the full assembly hierarchy). The BOM is the backbone of material requirements planning (MRP) — it drives what to order, when, and in what quantity.

C

Capacity Requirements Planning (CRP) — Production planning
The process of determining, for each work center, whether available machine and labor capacity is sufficient to execute a proposed production schedule (typically the output of MRP). CRP compares required hours against available hours per period and flags overloads before the schedule is released to the floor.
Cellular Manufacturing — Lean manufacturing
An arrangement of equipment and workstations, grouped by product family rather than by process function, so that a part can move through a complete sequence of operations within one compact "cell" without traveling long distances between departments. Cellular layouts reduce transport waste, work-in-process inventory, and lead time compared to traditional functional (process) layouts.
Control Chart — Statistical Process Control (SPC)
A time-ordered plot of a process measurement (such as a sample mean or range) against upper and lower control limits, typically set at ±3 standard deviations from the process mean. Points within the limits, showing no non-random patterns, indicate the process is in a state of statistical control (only common-cause variation present); points outside the limits or exhibiting run patterns (per Western Electric or Nelson rules) signal a special cause requiring investigation. Common chart types include X-bar/R, X-bar/S, p-charts, and c-charts.
Cp and Cpk — Process capability, SPC
Cp (process capability index) compares the width of the specification tolerance to the natural spread of the process (6 standard deviations), assuming the process is perfectly centered. Cpk accounts for how far the process mean is actually shifted from the center of the tolerance, so Cpk ≤ Cp always. A Cpk of 1.33 or higher is a common industry threshold for a "capable" process; a Cpk below 1.0 means the process will produce out-of-spec parts even when in statistical control.
Critical Path Method (CPM) — Project management
A project scheduling technique that models a project as a network of tasks with defined durations and dependencies, then identifies the longest chain of dependent tasks (the critical path) that determines the minimum possible project duration. Any delay on a critical-path task delays the entire project; tasks not on the critical path have "float" or "slack" — time they can be delayed without affecting the finish date.
Cycle Time — Manufacturing/lean
The total elapsed time from the start to the completion of one unit of production at a single process step, including both value-added work and any necessary handling. Distinguished from takt time (the required pace dictated by customer demand) — a station is a bottleneck if its cycle time exceeds takt time.

D

DMAIC — Six Sigma
The five-phase problem-solving framework used in Six Sigma improvement projects: Define the problem and project scope, Measure current performance with reliable data, Analyze the data to identify root causes, Improve the process by implementing and testing solutions, and Control the gains with standardized procedures and monitoring (often via SPC control charts). DMAIC is used for improving an existing process; DMADV/DFSS is the parallel framework for designing a new one.
Design of Experiments (DOE) — Statistics/quality engineering
A structured, statistically rigorous method for varying multiple input factors simultaneously (rather than one at a time) to determine their individual and interactive effects on a process output, using far fewer experimental runs than a full factorial test-everything approach. Common DOE structures include full factorial, fractional factorial, and Taguchi orthogonal array designs; the output is typically a regression model identifying which factors — and which factor interactions — matter most.

E

Economic Order Quantity (EOQ) — Inventory management
The order quantity that minimizes the total of ordering costs (fixed cost per order, independent of quantity) and holding costs (cost per unit per period to carry inventory), given a known, constant annual demand. The classic formula is EOQ = √(2DS/H), where D is annual demand, S is the fixed cost per order, and H is the annual holding cost per unit. EOQ assumes constant demand and lead time — real systems typically add a safety-stock buffer on top of the EOQ-driven reorder point to hedge against variability.
Ergonomics (Human Factors Engineering) — Occupational engineering
The scientific discipline concerned with designing tasks, tools, workstations, and systems to fit the physical and cognitive capabilities of the people who use them, reducing injury risk (especially musculoskeletal disorders) and improving performance. Common tools include the NIOSH lifting equation (for evaluating manual material-handling tasks), RULA/REBA postural assessment scores, and anthropometric design tables.

F

Facility Layout — Industrial engineering
The physical arrangement of departments, workstations, equipment, and material-flow paths within a facility. The four classic layout types are process (functional, grouping similar machines), product (line, arranged in the sequence of operations), cellular (grouped by part family), and fixed-position (product stays put, resources move to it — used for large items like aircraft or ships). Layout selection depends heavily on production volume and product variety.
FIFO (First-In, First-Out) — Inventory/lean
An inventory rule ensuring that the oldest units of stock are consumed or shipped before newer units, used both as an accounting method and, in lean manufacturing, as a physical material-flow discipline (a FIFO lane) to prevent stagnant inventory and control work-in-process levels between processes that cannot be directly connected by a continuous flow.
Failure Mode and Effects Analysis (FMEA) — Reliability/quality engineering
A systematic, tabular method for identifying every way a process or design can fail (failure modes), the effects of each failure, its likely causes, and current controls, then ranking each failure mode by a Risk Priority Number (RPN = Severity × Occurrence × Detection, each rated 1–10) to prioritize corrective action. Process FMEA (PFMEA) examines manufacturing steps; Design FMEA (DFMEA) examines the product design itself. Widely required in automotive (AIAG-VDA FMEA handbook) and aerospace supply chains.

G

Gage R&R (Gage Repeatability and Reproducibility) — Measurement systems analysis
A statistical study that quantifies how much of the total observed variation in a measurement comes from the measurement system itself (repeatability — variation from the same operator measuring the same part repeatedly, and reproducibility — variation between different operators) versus true part-to-part variation. A common acceptance guideline is that Gage R&R should consume less than 10% of total tolerance or variation for a measurement system to be considered acceptable for process control decisions; 10–30% is marginal, and above 30% the gage itself is unreliable for decision-making.
Gemba — Lean manufacturing
Japanese for "the actual place" — in lean practice, the shop floor or workplace where value is actually created. A "gemba walk" is a management practice of physically going to observe the work as it happens, rather than relying on reports, to understand problems and improvement opportunities firsthand.

H

Heijunka (Production Leveling) — Toyota Production System
The practice of smoothing production volume and product mix over time — rather than producing in large batches of one item — so that upstream processes and suppliers see a stable, predictable demand pattern instead of large swings. Heijunka is a prerequisite for effective kanban pull systems, since a wildly variable demand signal defeats the purpose of a steady pull.

I

ISO 9001 — Quality management
The international standard, published by the International Organization for Standardization, that specifies requirements for a quality management system (QMS) — covering leadership commitment, risk-based thinking, process control, document control, and continual improvement. ISO 9001 certification is issued by accredited third-party auditors and is frequently a customer or contractual requirement for suppliers in manufacturing, aerospace, and automotive supply chains (which layer additional sector-specific requirements on top, such as AS9100 for aerospace or IATF 16949 for automotive).
Inventory Turnover — Operations/finance
The ratio of cost of goods sold (or annual usage) to average inventory value over a period, indicating how many times inventory is fully cycled through per year. Higher turnover generally indicates leaner, more efficient inventory management and lower carrying cost, though turnover targets vary enormously by industry — a grocery distributor and an aerospace spares warehouse have very different reasonable benchmarks.

J

Jidoka — Toyota Production System
Often translated as "automation with a human touch," jidoka is the principle of building quality checks into a process — whether performed by a machine or a person — so that any abnormality or defect stops the process immediately rather than allowing bad product to continue downstream. Poka-yoke devices and andon systems are the practical mechanisms that implement jidoka.
Just-in-Time (JIT) — Toyota Production System
A production philosophy in which materials, components, and subassemblies arrive or are produced exactly when needed by the next process, in the exact quantity needed — no more, no less — minimizing inventory and the waste associated with carrying, storing, and handling excess stock. JIT depends on reliable, short-lead-time suppliers, small lot sizes, quick changeovers (SMED), and a pull signal such as kanban to trigger production or replenishment.

K

Kaizen — Lean manufacturing
Japanese for "change for the better" — the philosophy and practice of continuous, incremental improvement involving everyone in the organization, from line operators to executives. A "kaizen event" or "kaizen blitz" is a focused, typically 3-to-5-day workshop where a cross-functional team rapidly analyzes and improves a specific process, in contrast to large, slow, top-down improvement programs.
Kanban — Toyota Production System / pull systems
A visual signaling system (traditionally a physical card, but often a digital signal or empty bin today) that authorizes and triggers the movement or production of a specific quantity of material, only when downstream demand actually consumes it. Kanban implements a "pull" system — production is triggered by actual consumption rather than a forecast-driven "push" schedule — and is the core mechanism that enables just-in-time flow. Kanban boards (To Do / In Progress / Done) are also widely used outside manufacturing for visualizing knowledge work.

L

Lean Manufacturing — Manufacturing philosophy
A management philosophy, derived from the Toyota Production System, focused on maximizing customer value while minimizing waste (muda) in all its forms — overproduction, waiting, transport, over-processing, excess inventory, unnecessary motion, and defects (sometimes remembered as the "8 wastes" when non-utilized talent is added). Lean tools include value stream mapping, 5S, kanban, SMED, and kaizen, all aimed at creating smooth, continuous flow of value with minimal non-value-added activity.
Linear Programming (LP) — Operations research
A mathematical optimization technique for finding the best outcome (maximum profit, minimum cost) in a model whose objective function and constraints are all linear relationships among decision variables. LP is widely used in production planning, blending, transportation/distribution, and staffing problems, typically solved via the simplex method or interior-point algorithms in software such as Excel Solver, Gurobi, or CPLEX.

M

Materials Requirements Planning (MRP) — Production planning
A computerized planning system that starts from a master production schedule (what finished goods are needed, and when), explodes each product's bill of materials to determine component and raw-material requirements, nets those requirements against on-hand and on-order inventory, and offsets by lead time to generate a time-phased plan of what to order or produce and when. MRP II extends the concept to include capacity planning, scheduling, and financial integration; modern ERP systems (SAP, Oracle, etc.) build on MRP logic.
Mean Time Between Failures (MTBF) — Reliability engineering
The average operating time between failures for a repairable system, calculated as total operating time divided by number of failures over a given period. MTBF is a reliability metric, not a guarantee that any individual unit will run that long — it describes population-level failure behavior, most meaningful during a system's "useful life" period on the reliability bathtub curve (after infant-mortality failures and before wear-out failures dominate).
Mean Time to Repair (MTTR) — Reliability/maintenance engineering
The average time required to diagnose and repair a failed component or system and return it to operation, including detection, diagnosis, repair, and verification time. MTTR is a key input to availability calculations (Availability = MTBF / (MTBF + MTTR)) and drives maintainability design decisions such as modular components and quick-disconnect fasteners.
Muda, Mura, Muri — Toyota Production System
The three categories of waste targeted by lean manufacturing. Muda is non-value-added waste in a process (the classic 8 wastes). Mura is unevenness or variability in workload or demand, which itself creates waste. Muri is overburden — pushing a person or machine beyond its reasonable capacity, causing breakdowns, defects, or injury. Lean practitioners address mura and muri (through heijunka and standardized work) as root causes that generate muda downstream.

O

Overall Equipment Effectiveness (OEE) — Manufacturing performance metric
A composite metric equal to Availability × Performance × Quality, expressed as a percentage, that measures how effectively a piece of equipment or a production line is utilized against its theoretical maximum. Availability captures downtime losses, Performance captures speed losses (running slower than ideal cycle time), and Quality captures losses from defective or scrap output. A widely cited "world-class" OEE benchmark is 85%, though realistic targets vary substantially by industry and process type.

P

Pareto Analysis — Quality engineering
A prioritization technique based on the Pareto principle ("80/20 rule") that ranks causes or categories of a problem by frequency or impact, typically displayed as a Pareto chart — a bar chart in descending order with a cumulative-percentage line overlaid — to visually identify the "vital few" contributors responsible for most of the effect, focusing improvement effort where it will have the greatest impact.
PDCA (Plan-Do-Check-Act) — Quality management
A four-step iterative cycle for continuous improvement, popularized by W. Edwards Deming: Plan a change and predict its result, Do (implement) the change on a small scale, Check the results against the prediction, and Act by standardizing the change if successful or adjusting the plan if not. PDCA is the conceptual ancestor of DMAIC and underlies most structured kaizen activity.
PERT (Program Evaluation and Review Technique) — Project management
A project scheduling technique, closely related to CPM, that models task durations probabilistically using optimistic, most-likely, and pessimistic time estimates to compute an expected duration and variance for each task and for the overall project, useful when task durations are uncertain rather than fixed and known.
Poka-Yoke — Toyota Production System
Japanese for "mistake-proofing" — a design feature, fixture, or procedure that makes it physically impossible (or immediately obvious) to perform an operation incorrectly, such as a connector that only fits one way, a fixture that will not close if a part is missing, or a counting sensor that flags a short kit. Poka-yoke devices are the physical implementation of jidoka's "stop and fix" principle at the point of error rather than relying on inspection to catch it later.
Predictive Maintenance — Maintenance engineering
A maintenance strategy that uses condition-monitoring data — vibration analysis, oil analysis, infrared thermography, ultrasonic testing — to detect early signs of equipment degradation and schedule maintenance only when evidence indicates it is actually needed, in contrast to preventive maintenance (fixed time/usage intervals regardless of condition) or reactive maintenance (repair after failure). Predictive maintenance typically reduces both unplanned downtime and unnecessary maintenance labor compared to a purely time-based schedule.

Q

Queuing Theory — Operations research
The mathematical study of waiting lines — modeling arrival rates, service rates, number of servers, and queue discipline to predict metrics such as average wait time, queue length, and server utilization. The Kendall notation (e.g., M/M/1) classifies queuing models by arrival distribution, service distribution, and number of servers; queuing theory underlies capacity planning for call centers, toll booths, hospital emergency departments, and manufacturing buffer sizing.

R

Reliability — Reliability engineering
The probability that a system or component will perform its intended function without failure for a specified period under stated operating conditions. Reliability is typically modeled with the exponential distribution (constant failure rate, common in the useful-life period) or the Weibull distribution (which can model increasing or decreasing failure rates, useful for wear-out or infant-mortality behavior across a component's life).
Root Cause Analysis (RCA) — Quality/problem-solving
A structured investigation method for identifying the underlying, fundamental cause of a problem rather than treating its symptoms, so that a corrective action prevents recurrence rather than merely suppressing the immediate effect. Common RCA tools include the "5 Whys" (iteratively asking why a problem occurred until the root cause surfaces) and the fishbone/Ishikawa diagram (categorizing potential causes under headings such as Man, Machine, Method, Material, Measurement, and Environment).

S

Safety Stock — Inventory management
Extra inventory held above expected demand during lead time, sized to buffer against variability in demand or supplier lead time and to protect a target service level (probability of not stocking out) against that variability. Safety stock is commonly calculated from the standard deviation of demand/lead time and a service-level factor (a z-score from the normal distribution), and it sits on top of — not instead of — cycle stock ordered via the EOQ or reorder-point logic.
Single-Minute Exchange of Die (SMED) — Lean manufacturing
A methodology, developed by Shigeo Shingo, for radically reducing equipment changeover/setup time — ideally to under 10 minutes ("single minute" meaning single-digit minutes) — primarily by converting internal setup steps (which require the machine to be stopped) into external setup steps (which can be prepared while the machine is still running) and then simplifying and standardizing whatever internal steps remain. Fast changeovers make small batch sizes economically viable, which is a prerequisite for heijunka and JIT.
Six Sigma — Quality management
A data-driven quality methodology aimed at reducing process variation to the point where defects occur at a rate of no more than 3.4 per million opportunities (a "six sigma" level of process capability, referring to fitting six standard deviations between the process mean and the nearest specification limit). Six Sigma projects are structured around the DMAIC framework and use a belt-based practitioner hierarchy (Yellow, Green, Black, Master Black Belt) denoting increasing statistical and project-leadership expertise.
Standard Work — Lean manufacturing
The current best-known, documented method for performing a task — specifying the sequence of steps, the takt-time-driven cycle time, and the standard quantity of in-process material — used both as a training tool and as a stable baseline against which kaizen improvements can be measured. Standard work is meant to be a living document, updated whenever a kaizen event finds a better method, not a rigid, permanent rulebook.
Statistical Process Control (SPC) — Quality engineering
The use of statistical methods, primarily control charts, to monitor and control a process in real time, distinguishing normal, expected variation (common cause) from abnormal variation (special cause) that signals something in the process has genuinely changed and needs investigation. SPC is a monitoring technique for a process already deemed capable (see Cp/Cpk); its goal is to keep a process stable, not to make it capable in the first place.

T

Takt Time — Lean manufacturing
The rate at which a product must be completed to meet customer demand, calculated as available production time per period divided by customer demand for that period. Takt time is a target pace, not a measured cycle time — a line is balanced when the cycle time of each station is at or just under takt time, with no single station taking longer (which would create a bottleneck) and no station idling excessively (which wastes capacity).
Theory of Constraints (TOC) — Operations management
A management philosophy, developed by Eliyahu Goldratt, holding that any system has at least one constraint (bottleneck) limiting its overall throughput, and that system-wide improvement comes from systematically identifying and elevating that constraint using the Five Focusing Steps, rather than optimizing every resource independently (which Goldratt argued often makes overall performance worse, not better).
Total Productive Maintenance (TPM) — Maintenance/lean
A maintenance philosophy that makes equipment reliability everyone's responsibility — not just the maintenance department's — by training operators to perform basic daily care, cleaning, and inspection ("autonomous maintenance") alongside a structured planned-maintenance program, aiming to eliminate the "six big losses" that reduce OEE (breakdowns, setup/adjustment, minor stops, reduced speed, startup defects, and process defects).

V

Value Stream Mapping (VSM) — Lean manufacturing
A visual mapping technique that documents every step — both value-added and non-value-added — required to bring a product or service from raw material/request to the customer, capturing process time, wait time, inventory levels, and information flow at each step. A "current state" map identifies waste and improvement opportunities; a "future state" map defines the target process design, and the gap between them becomes the improvement roadmap.
Variation, Common Cause vs. Special Cause — Statistical Process Control
Common-cause variation is the natural, inherent "noise" always present in a stable process — small, random fluctuations from many minor sources that are expensive or impossible to eliminate individually. Special-cause variation comes from an identifiable, assignable source outside the normal system — a broken tool, a wrong material lot, an untrained operator — and should be investigated and corrected whenever a control chart signals its presence. Confusing the two (reacting to every common-cause blip as if it were special) is called "tampering" and typically increases process variation rather than reducing it.

W

Work-in-Process (WIP) — Manufacturing/lean
Partially completed products that have entered the production process but have not yet been finished and moved to finished-goods inventory. Excess WIP is one of the classic lean "8 wastes" — it ties up cash, hides quality problems (defects sit unfound in a large WIP buffer), and extends lead time; WIP limits (a core practice in kanban and pull systems, and later formalized in the Kanban Method for knowledge work) cap how much can be in progress at once to keep flow smooth and surface bottlenecks quickly.