Begrippen Organization Management & Logistics

Operations and supply chain management (OSCM): Designing, running and improving the systems through which a firm creates and delivers its main products and services.
Process: One or more activities that transform inputs into outputs; the means by which a firm carries out its strategy.
Efficiency: Doing something at the lowest possible cost; in process analysis, actual output divided by standard output.
Order winners and order qualifiers: Order winners distinguish a firm from competitors; order qualifiers are the minimum characteristics needed to be considered at all.
Operations and supply chain strategy: Setting broad policies and plans for using the firm's resources to best support its long-term competitive strategy.
Concurrent engineering: Developing a product with specialists from several functions working in parallel rather than sequentially.
Critical path: The longest sequence of connected activities in a project network; it determines the shortest project duration.
Slack: The time an activity can be delayed without delaying the whole project.
Capacity: The output a system can achieve over a specific period of time.
Best operating level: The capacity level for which a process was designed and where average unit cost is minimal.
Capacity cushion: Capacity in excess of expected demand, expressed as a percentage of demand.
Economies of scope: Producing several products together more cheaply than producing each separately, thanks to flexible processes.
Bottleneck: The resource that limits the capacity of the whole process.
Cycle time: The average time between completions of successive units.
Flow time: The average time a unit takes to move through the entire process, including waiting.
Little's law: In steady state, inventory equals throughput rate times flow time.
Blocking and starving: Blocking: a stage stops because it cannot pass on its output; starving: a stage stops because it has no work.
Design quality versus conformance quality: Design quality is the inherent market value of a product; conformance quality is the degree to which its specifications are met.
Cost of quality (COQ): Costs of prevention, appraisal, internal failure and external failure related to quality.
DPMO: Defects per million opportunities: defects divided by (opportunities per unit times units), times one million.
Poka-yoke: A fail-safe device or procedure that prevents an error or gives immediate feedback.
Muda: Japanese term for waste: anything that does not add value from the customer's perspective.
Kanban: A card or signal that authorizes moving or producing a container of parts in a pull system.
Heijunka (uniform plant loading): Producing the same small mix of products every day to smooth the load on the system.
Value stream: All value-adding and non-value-adding activities needed to design, order and deliver a product.
Decoupling point: A point in the supply chain where inventory separates stages, allowing them to work independently.
Exponential smoothing: A forecast method in which the new forecast equals the previous forecast plus a fraction alpha of the last error.
Mean absolute deviation (MAD): The average of absolute forecast errors; about 0.8 times the standard deviation for normal errors.
Tracking signal: The running sum of forecast errors divided by MAD; values far from zero indicate bias.
Delphi method: A qualitative forecasting method with anonymous experts who answer several rounds of questionnaires with summarized feedback.
Independent versus dependent demand: Independent demand is unrelated to other items and must be forecast; dependent demand follows from the need for a higher-level item.
Economic order quantity (EOQ): The order size that minimizes total ordering and holding costs: sqrt(2DS/H).
Safety stock: Inventory carried in addition to expected demand to protect against variability.
Inventory position: On-hand plus on-order minus backordered inventory.
ABC classification: Grouping items by annual dollar volume so that the most valuable items receive the tightest control.
Operations: In the book's specific meaning: the manufacturing and service processes that transform a firm's resources into the products customers want, such as a car plant, a call center or an emergency room. They are distinguished from supply chain processes, which move material and information to and from these operations.
Supply chain: The processes that move information and material to and from a firm's operations, on the input side (supplies to plants) and the output side (products to customers), including logistics, warehousing and storage. The same product can use very different supply chains: a custom pair of jeans flown overnight versus a low-cost pair shipped by container from Asia.
Supply chain process categories: From a producer's perspective, supply chain processes fall into five groups: planning (meeting anticipated demand with available resources), sourcing (selecting suppliers, pricing, receiving and paying), making (producing the good or service), delivering (carriers, warehouses, customer orders, invoicing) and returning (taking back worn-out, defective or excess products, after-sales support).
Service package: The bundle that makes up a service offering: the supporting facility, facilitating goods, explicit services and implicit services such as atmosphere, waiting time and privacy. Services are specified as such a package of features, whereas goods are defined by physical specifications.
Goods-services continuum: The idea that almost every offering lies on a scale running from pure goods, through core goods and core services, to pure services. Producers of pure goods often move along this scale by adding services, to escape low-margin commodity competition.
Product-service bundling: Building service activities such as maintenance, spare parts, training and systems design into the product offering. It tends to raise revenue, but research cited in the book shows that bundling firms often earn a lower profit as a percentage of revenue than focused firms, because the extra service investment is costly.
Effectiveness: Doing the things that create the most value for the customer. It can conflict with efficiency: a grocery store keeps costs low with as few cashiers as possible, while effectiveness requires enough open checkout lines so that customers do not have to wait.
Value: How attractive a product is compared with what it costs: value = quality / price. A better product at the same price gives more value, and a better product at a lower price gives even more.
Operations effectiveness: Performing the core business processes in a way that best implements the firm's strategic priorities at minimum cost. Initiatives aimed at it, such as quality programs, process redesign, planning and control systems and new technology, can show results within 12 to 24 months.
Competitive dimensions: The dimensions on which an operation can compete: cost or price, quality (design and process quality), delivery speed, delivery reliability (delivering on or before the promised date), coping with changes in demand, flexibility and new-product introduction speed, and other product-specific criteria such as technical support or environmental impact.
Operations focus and trade-offs: The logic that one operation cannot excel on every competitive dimension at once, so management must choose the critical performance parameters and concentrate resources on them. Examples: focusing on delivery speed limits product variety, low cost conflicts with speed and flexibility, and an airline cannot both serve meals and turn planes around fast.
Straddling: Trying to match a competitor by adding features, services or technologies on top of existing activities while keeping the current position, as when BMW targets both the luxury and the performance segment. Because trade-offs still apply, straddling firms often lose market share in both segments.
Activity-system map: A diagram showing how a firm's higher-order strategic themes are implemented through clusters of tightly linked activities. Competitive advantage comes from the fit among these activities, as at IKEA, where self-service displays, self-designed ready-to-assemble furniture and a self-serve warehouse reinforce a low-cost strategy.
Supply chain risk: The chance that something interrupts a firm's ability to keep delivering its products or services. Coordination risks come from daily operations and are covered with safety stock, extra lead time or overtime; disruption risks come from disasters such as earthquakes, strikes or pandemics, whether natural or caused by people.
Risk management framework: A three-step approach to supply chain risk: identify the sources of potential disruptions, assess their impact by probability and consequence, and develop plans for the high and moderate risks. The mitigation strategies are avoid, accept, transfer (for low-probability, high-impact risks, for example insurance or outsourcing) and limit, the most common one.
Productivity: Outputs divided by inputs, a relative measure that only means something when compared with similar operations or the same operation over time. A partial measure uses one input (output per labor hour), a multifactor measure a group of inputs (labor + material + overhead) and a total measure all outputs over all inputs.
Triple bottom line: Evaluating a firm on three criteria, also called People, Planet, Profit: social responsibility (fair practices toward labor and community), economic prosperity (a competitive return for shareholders and lasting benefit to society) and environmental stewardship (reducing the ecological footprint). It underlies a sustainable strategy that considers all stakeholders.
Core competency: The one thing a firm does better than its competitors and that yields a long-term competitive advantage, such as engines for Honda. It has three characteristics: it gives potential access to a wide variety of markets, it increases perceived customer benefits, and it is hard for competitors to imitate.
Generic product development process: Six phases: planning (closing with a mission statement), concept development, system-level design, detail design, testing and refinement with preproduction prototypes, and the ramp-up to full production. Marketing, design and manufacturing take part throughout. This standard process suits market-pull products; technology-push, platform and high-risk products follow adapted versions.
Quality function deployment (QFD): An interfunctional approach (marketing, engineering and manufacturing) to capture the voice of the customer. Customer requirements are gathered and weighted by importance, and customers compare the product with competitors. Its matrix, the house of quality, translates customer requirements into technical characteristics and targets.
Value analysis / value engineering (VA/VE): Simplifying products and processes to get equal or better performance at lower cost while keeping all customer functions. VA is applied to products already in production (cost reduction), VE before production starts (cost avoidance). Typical questions: are there unnecessary features, can parts be combined, can weight be cut?
Design for manufacturing and assembly (DFMA): Reducing production cost mainly by cutting the number of separate parts. Each part is tested: does it move relative to other parts, must it be of a different material, must it be separate for disassembly? If all answers are no, it can be combined or eliminated. In the motor drive example, 19 parts were reduced to 7 (theoretical minimum 4).
Ecodesign: Integrating environmental considerations into product design on three levels: the complete life cycle from extraction to waste treatment, the product as a system including consumables, packaging and energy, and a multicriteria approach that avoids shifting impact between categories such as resource depletion and toxicity.
Complexity and divergence: Two dimensions for designing services. Complexity is the number of steps and the possible actions at each step; divergence is how much the customer/server interaction can vary at each step. A family restaurant can lower both (buffet, self-service) or raise both (maitre d', table-side preparation).
Base-case financial model: An estimate of the timing and size of a development project's cash flows (development, ramp-up, marketing and support, production cost, sales revenue), discounted to a net present value. PV = CF / (1 + r)^t and NPV is the sum of all period PVs; past spending is a sunk cost and is ignored.
Sensitivity analysis: A what-if analysis that changes one factor of the base-case model, such as development cost, development time, sales volume or unit price, and recalculates the NPV. In the CI-700 printer example, a 20 percent lower development cost raised the NPV by about 11.8 percent.
Project management: Planning, steering and controlling people, equipment and materials so that a project meets its technical, cost and time targets. A project is a set of linked jobs aimed at one major result over a longer period. Soft skills such as motivating people and resolving conflict count as much as planning techniques.
Pure project: A project structure with its own dedicated team that works on the project full time. Pros: the project manager has complete authority, members report to one boss, communication is quick and commitment is strong. Cons: resources are duplicated, functional expertise erodes, and ending the project gets delayed because members worry about what comes next.
Functional project: A project structure in which the project is housed in a functional division and members stay in their own units. Expertise stays in the function, members can work on several projects and have a home afterwards, but non-functional aspects are shortchanged, motivation is weak and client needs are handled slowly.
Matrix project: A project structure in which the project manager decides what is done and when, while functional managers decide who does it and with which technology. It improves communication between functions and limits duplication, but members have two bosses and the PM needs strong negotiating skills. A matrix can be weak, balanced or strong.
Work breakdown structure (WBS): The hierarchy of a project's tasks, subtasks and work packages, derived from the statement of work (objectives, work, schedule, budget, milestones, reports). A work package is a group of activities assignable to one organizational unit, with its own schedule, budget and measures.
Earned value management (EVM): A control method that combines scope, schedule and cost, because comparing actual cost with budget alone ignores how much work is done. BCWS (planned value) = sum of planned percent complete x activity budget, BCWP (earned value) = sum of actual percent complete x activity budget, and AC is the actual cost.
Schedule and cost variance (SV, CV): EVM measures: SV = BCWP - BCWS and SPI = BCWP / BCWS for the schedule, CV = BCWP - AC and CPI = BCWP / AC for cost. Variances above 0 or indices above 1 are good. In Example 4.1, SPI = 0.8 means behind schedule and CPI = 0.89 means over budget.
Forward and backward pass: The CPM calculations of activity times. Forward pass: ES = latest EF of the predecessors and EF = ES + t. Backward pass: LF = earliest LS of the successors and LS = LF - t. Slack then equals LS - ES, and activities with zero slack lie on the critical path.
Three time estimates (PERT): For uncertain activity times: expected time ET = (a + 4m + b) / 6 with optimistic a, most likely m and pessimistic b (beta distribution), and variance ((b - a) / 6)^2. Path time is approximately normal, so Z = (D - TE) / √(sum of variances on the critical path) gives the probability of finishing by due date D.
Crashing: Shortening a project by spending more on activities. Direct costs rise as the project gets shorter while indirect costs fall; the goal is the duration with minimum total cost. Cost per day = (CC - NC) / (NT - CT); shorten the cheapest critical activity one day at a time, and parallel critical paths together.
Capacity planning horizons: Capacity decisions differ by time range. Long range (more than a year) covers buildings, equipment and facilities and needs top management approval. Intermediate range (the next 6 to 18 months) covers hiring, layoffs, minor equipment and subcontracting. Short range (under a month) covers scheduling, overtime and personnel transfers.
Capacity utilization rate: Capacity used divided by the best operating level, in percent, with both in the same units and time period. A plant designed for 600 units per day that produces 480 units per day has a utilization of 480 / 600 = 80 percent.
Economies and diseconomies of scale: As a plant grows and volume rises, average unit cost falls, because equipment with twice the capacity does not cost twice as much and support resources are better used. Beyond some size, diseconomies appear: heavy discounting to keep the plant busy, high maintenance of a few large machines, or transport costs, as for a cement plant.
Focused factory: A facility that concentrates on a limited set of production objectives, such as one product or product group, instead of trying to excel on every dimension. It can be realized as a plant within a plant (PWP): separate areas under one roof, each with its own organization, equipment, workforce policies and control.
Capacity flexibility: The ability to raise or lower output quickly or to shift capacity between products. It comes from flexible plants (zero-changeover-time plants with movable equipment), flexible processes (flexible manufacturing systems and easily set up equipment) and flexible workers with multiple skills.
Maintaining system balance: Stages of a process rarely have identical best operating levels, and variability creates imbalance. When changing capacity, remedies are adding capacity at bottlenecks (overtime, leasing, subcontracting), holding buffer inventory before the bottleneck, or duplicating the facilities on which another department depends.
Decision tree: A schematic model of a sequence of decisions (squares) and chance events (circles, with probabilities on their branches). It is solved from right to left: take the expected value at chance nodes and keep only the best branch at decision nodes. With long horizons, use present values. In Example 5.2, doing nothing ($703,750) beats expanding and moving.
Rate of service utilization: ρ = mean arrival rate / mean service rate. The best operating point is near 70 percent; in the critical zone service quality declines, and above 100 percent the queue keeps growing. The optimal rate is low for uncertain, high-stakes services (emergency rooms) and close to 100 percent for predictable services without customer contact (commuter trains).
Utilization: The time a resource is actually activated divided by the time it is available. In practice setup time is often excluded, so always ask how the metric was calculated. In the slot machine example, a 10-second machine at 50 percent utilization earns the casino $216 per day versus $144 for a 15-second machine.
Buffer: A storage area between process stages that lets the stages work independently; without buffers, blocking and starving occur. Buffers add waiting time: a six-station paced line with a 30-second cycle has a flow time of 3 minutes, which grows to 8 minutes with 10 always-full buffer positions.
Make-to-order, make-to-stock and hybrid processes: Make-to-order builds a product from raw materials in response to a specific order (traditional Wendy's burger). Make-to-stock serves customers from finished goods inventory (traditional McDonald's holding bin). A hybrid process stocks a generic product and finishes it to order, as with Burger King's conveyor-broiler and the current McDonald's system.
Pacing: Coordinating movement through a process with a timing mechanism, such as a line that moves every 45 seconds. The required cycle time follows from available production time / required output, for example 420 x 60 seconds / 1,000 cars = 25.2 seconds per car.
Operation time: Setup time plus run time for a batch. Setup time is the time to prepare a machine for a particular item; run time is time per unit x batch size. A bottle filler with a 30-minute setup and 2 seconds per bottle needs 1,800 + 20,000 = 21,800 seconds (363.33 minutes) for 10,000 bottles.
Process velocity (throughput ratio): In the chapter's definition, value-added time divided by flow time, which shows what share of the time a unit spends in the process is actual work. A paced line with 3 minutes of work and a flow time of 8 minutes, because of full buffers, has a process velocity of 3 / 8 = 0.375.
Inventory turn: Cost of goods sold divided by average inventory value, a better basis for comparing firms than the total inventory value. Its inverse gives days-of-supply: 6 turns per year means 1/6 x 365 = about 61 days of supply.
Job enrichment: Making specialized work more interesting within job design, which specifies the work activities of an individual or group. Horizontal enlargement adds more or more varied tasks; vertical enlargement adds planning, organizing and inspecting one's own work. Enrichment strongly improves quality, while its effect on productivity is less predictable.
Work measurement: Setting time standards to schedule work, motivate and measure workers, bid on contracts and benchmark. Direct methods are time study (stopwatch) and work sampling (random observations, for infrequent or long-cycle work); indirect methods are predetermined motion-time data systems and elemental data. Critics point to unattainable standards and rate cutting.
Total quality management (TQM): Running the whole organization so that it performs excellently on every product and service dimension the customer cares about. In practice it has two aims: design the product or service carefully, and make sure the organization's systems deliver that design consistently. The Baldrige Award (1987) honours firms with outstanding quality systems.
Quality gurus (Crosby, Deming, Juran): Crosby defined quality as conformance to requirements, aimed at zero defects and held that quality is free. Deming stressed reducing variability through continuous improvement and ceasing mass inspection (14 points). Juran defined quality as fitness for use and argued quality is not free: there is an optimum.
Dimensions of design quality: Six aspects of how well a product is designed: performance (its main characteristics), features (extra touches), reliability and durability (consistency, chance of failure, lifespan), serviceability (how easy it is to repair), aesthetics (how it sounds, feels and looks) and perceived quality (reputation and track record).
Quality at the source: The principle that the person doing the work is responsible for meeting the specifications. In lean production it means doing it right the first time, with workers allowed to stop the line when there is a problem.
Six Sigma: Both a statistical goal of no more than 3.4 defects per million and a philosophy and program for removing defects by reducing process variation, which is seen as the enemy of quality. A defect is anything outside the customer's specification limits. Roles include champions, master black belts, black belts (team leaders) and green belts.
DMAIC: The Six Sigma improvement cycle: Define (customers, priorities, project and critical-to-quality characteristics), Measure (how to measure the process and current defect levels), Analyze (most likely causes and key variables), Improve (remove causes, set acceptable ranges) and Control (keep the key variables within range).
Failure mode and effect analysis (FMEA): A structured analysis in which a risk priority number (RPN) is computed for each possible failure mode, based on its occurrence, severity and detection. The failure modes with the highest RPNs are tackled first. It is one of the analytical tools used in Six Sigma projects.
Shingo system: Shigeo Shingo's fail-safe approach: errors are inevitable, but defects can be prevented through immediate feedback and correction, whereas statistical quality control only detects defects afterwards. Inspection covers 100 percent of items as a successive check (next person), self-check (the worker) or source inspection, supported by poka-yokes.
ISO 9000: The international standard for quality management systems: firms document and implement their quality system and are verified by an accredited third party. It rests on seven principles, including customer focus, leadership, process approach and continual improvement. ISO 14000 covers environmental management in the same way.
Lean production: A set of linked activities aimed at producing large volumes at high quality while keeping stocks of raw materials, work in process and finished goods as low as possible. Parts arrive exactly when needed and nothing is made before there is demand. Lean comes from Toyota's JIT system and works best at steady volumes, which is why sudden swings such as during COVID-19 led to shortages.
Pull system: A system in which production is triggered by actual demand: a sale pulls a replacement from final assembly, which pulls from the upstream station, and so on back to raw materials. It requires high quality at each stage, strong vendor relations and fairly predictable demand.
Toyota Production System (TPS): Toyota's system built on two philosophies: elimination of waste (muda) and respect for people, such as job security, cooperative company unions and long-term supplier partnerships. It names seven types of waste: overproduction, waiting time, transportation, inventory, processing, motion and product defects.
Lean supply chain: Applying lean to the whole chain to maximize value from the end customer's view. Components are lean suppliers, lean procurement, lean manufacturing, lean warehousing, lean logistics (pooled orders, cross docking, backhaul minimization) and lean customers. Benefits are responsiveness, less obsolescence, shorter flow time and lower cost.
Value stream mapping (VSM): A flowcharting tool that shows material and information flows and where value is or is not added. Process boxes list cycle time, changeover time, lot size and uptime, triangles show inventory, and a time line compares value-added time with total lead time. A current state map leads to a future state map, with kaizen bursts marking improvement projects.
Group technology: Grouping similar parts into families that are made in a dedicated cell, instead of moving jobs between specialized departments. It reduces movement, waiting time, inventory and the number of employees needed, but requires flexible workers.
Internal and external setup: Splitting setup activities into internal ones, done while the machine is stopped, and external ones, done while it is still running. Minimizing setup times lowers setup cost and therefore the optimal lot size, which supports the small lots of lean production.
Number of kanban card sets: k = D x L x (1 + S) / C, rounded up, with D the average demand per period, L the replenishment lead time, S safety stock as a fraction of lead time demand and C the container size. Example 14.1: 8 x 4 x 1.1 / 10 = 3.52, so 4 sets. The number of containers caps work-in-process.
Lean services: Applying lean in services, which is harder because of uncertainty in task times, demand and customers' roles. Successful techniques include problem-solving groups, better housekeeping, clearer process flows, leveling the facility load (peak pricing, take-a-number), eliminating unnecessary activities and demand-pull scheduling.
Components of demand: Demand consists of six components: average demand, trend, seasonal element, cyclical elements (such as elections, war or the economy), random variation (what remains unexplained) and autocorrelation (a value is correlated with its own past values). Trends can be linear, S-curve, asymptotic or exponential.
Simple moving average: F(t) = (A(t-1) + A(t-2) + ... + A(t-n)) / n, the average of the last n periods. It removes random fluctuation when demand has no trend or seasonality. Fewer periods follow changes closely but fluctuate more; more periods smooth more but lag the trend. A drawback is that all data must be stored.
Weighted moving average: F(t) = w(1)A(t-1) + w(2)A(t-2) + ... + w(n)A(t-n), with the weights summing to 1. The most recent period usually gets the highest weight, for example 0.4, 0.3, 0.2 and 0.1, but with seasonal data the same season last year may deserve more weight.
Exponential smoothing with trend (FIT): Adds a trend term to correct the lag of single smoothing: F(t) = FIT(t-1) + α(A(t-1) - FIT(t-1)), T(t) = T(t-1) + δ(F(t) - FIT(t-1)) and FIT(t) = F(t) + T(t). Example 18.1: FIT = 110, T = 10, α = 0.2, δ = 0.3, actual 115 gives F = 111, T = 10.3 and FIT = 121.3.
Linear regression: Fitting a line Y = a + bt by least squares, minimizing the sum of squared vertical distances. It suits long-term forecasts of aggregate demand when data are roughly linear, for both time series and causal forecasting. In Example 18.2, quarter 13 is forecast as 441.67 + 359.6 x 13 = about 5,116.
Seasonal factor: Average demand in a season divided by average demand over all seasons, used when decomposing a time series. With multiplicative seasonality, the usual case, the forecast is (expected total / number of seasons) x factor. Example 18.3: seasonal demand of 200, 350, 300 and 150 out of 1,000 gives factors 0.8, 1.4, 1.2 and 0.6.
Mean absolute percent error (MAPE): The average absolute error as a percentage of actual demand: MAPE = (sum of |A(t) - F(t)| / A(t)) / n x 100 percent. Because it is a relative measure, it allows comparison of forecast accuracy between products with very different demand levels.
Causal relationship forecasting: Forecasting with independent variables other than time, which must be leading indicators known before demand occurs. In Example 18.5, carpet sales Y = 6,698.49 + 344.22 X with X the number of housing permits, so 25 permits give about 15,304 square yards. Multiple regression uses several variables but needs much external data.
Collaborative planning, forecasting and replenishment (CPFR): A web-based tool to coordinate forecasting, production, purchasing and replenishment among supply chain partners. Its five steps are a partnership agreement, joint business planning, developing demand forecasts, sharing forecasts with exception notices, and inventory replenishment. The main barrier is lack of trust.
Holding (carrying) cost: The cost of keeping items in stock: storage, handling, insurance, pilferage, obsolescence, taxes and the opportunity cost of capital. High holding costs favor small inventories and frequent orders. In the EOQ model it appears as H, the annual holding cost per unit.
Ordering, setup and shortage costs: Ordering cost is the managerial and clerical cost of preparing and tracking orders; setup cost is the cost of changing production to another product. Both favor large, infrequent orders or lots, which is why JIT aims to cut setup cost. Shortage cost arises from stockouts (lost sales) and backorders and favors extra stock.
Single-period model (newsvendor): For one-time purchases such as newspapers or fashion items: stock up to where P <= Cu / (Co + Cu), with Cu the cost of underestimating demand and Co the cost of overestimating. Order quantity = mean + z x σ. A paper costing $0.20 and selling for $0.50 gives P = 0.30 / 0.50 = 0.6, so 93 papers at mean 90 and σ 10.
Fixed-order quantity model (Q-model): An event-triggered inventory system: a constant quantity Q is ordered whenever the inventory position drops to the reorder point R. It needs perpetual recordkeeping of every withdrawal and addition, leads to a smaller inventory than the P-model, and suits expensive, critical or important items.
Fixed-time period model (P-model): A time-triggered inventory system: stock is counted at each review period T and a varying quantity is ordered, q = d(T + L) + zσ(T+L) - I, with I the current inventory position. It needs more safety stock, because it must also protect during the review period, and suits lower-cost items or items ordered together from one supplier.
Reorder point: The inventory position at which a new order is placed in a Q-model. With certain demand R = d x L; with uncertain demand R = d x L + zσL, where σL = √L x σd and z sets the service level (z = 1.64 for 95 percent). Example 20.4: R = 360 + 1.64 x 17.15 = 388 units.
Price-break model: An EOQ variant for quantity discounts, with H = iC. Starting at the lowest price, compute the EOQ per price until one is feasible within its range; then compare its total cost with the total cost at each lower-price break quantity. In Example 20.8, ordering 1,000 units at $3.90 ($39,590) beats the feasible EOQ of 667 at $4.50.
Cycle counting: Counting inventory items frequently instead of once or twice a year, to protect inventory accuracy (agreement between records and physical stock). Counts can be triggered by low or zero balances, discrepancies, activity levels or ABC importance. Recommended tolerances are plus or minus 0.2 percent for A, 1 percent for B and 5 percent for C items.