What you will learn in Demand Forecasting Best Practices
Chapter by chapter
Chapter 13: What do you review? ABC XYZ segmentations and other methods
What you learn in chapter 13 →Concepts explained in this book
- demand forecastA demand forecast is a prediction of future demand for a product, service, or item over a specified period.
- forecast accuracyForecast accuracy is the degree to which forecast values agree with the corresponding actual demand values.
- forecast value added frameworkThe forecast value added framework is a method for evaluating each step or contributor in a forecasting process by comparing its forecast performance with that of the preceding step.
- machine learningMachine learning is a computational approach in which models learn patterns from data to make predictions or decisions.
- biasForecast bias is the systematic tendency of forecasts to overestimate or underestimate actual demand.
- benchmarkA benchmark is a reference forecasting model or comparison value used to judge the performance of another forecasting model or process.
- forecast baselineA baseline forecast is a forecast produced before teams apply additional human judgments or modifications.
- unconstrained demandUnconstrained demand is what customers initially request in terms of product, quantity, and delivery date without limitations imposed by supply or inventory.
- value-weighted forecasting metricA value-weighted forecasting metric is a forecast-performance measure that weights products or errors according to their economic or operational importance.
- cognitive biasA cognitive bias is a systematic deviation in judgment caused by mental shortcuts, assumptions, or other distortions in human reasoning.
- market demandDemand is the quantity of a product or service that customers seek to purchase under specified conditions, including conditions observed in the past.
- salesSales are completed transactions or the quantities of products or services sold during a specified period.
- mean absolute errorMean absolute error is the arithmetic mean of the absolute differences between forecasted and corresponding actual values.
- demand planning processDemand planning is the process of estimating future customer demand and incorporating that estimate into operational plans.
- forecasting qualityThe overall degree to which forecasts meet desired performance criteria, including accuracy and bias.
- independence of work and judgmentIndependence of work and judgment is the practice of having contributors form assessments separately before group influence occurs.
- mindset diversityMindset diversity is the inclusion of contributors with different backgrounds, perspectives, information sources, and reasoning styles.
- forecastingA forecast is a prediction about a future quantity or condition based on available information.
- constrained salesConstrained sales are recorded sales quantities limited by supply, inventory, capacity, or other constraints rather than reflecting all underlying customer demand.
- forecast horizonA forecasting horizon is the future time span covered by a forecast.
- anchoring biasAnchoring bias is the tendency for an initial number or idea to influence subsequent estimates or judgments, even when it is unrelated or weakly relevant.
- external factorAn external driver is a factor outside the demand process that influences the level or pattern of demand.
- objective alignmentObjective alignment is the condition in which contributors share goals that support unbiased estimation rather than strategically distorted predictions.
- forecast errorForecast error is the difference between a forecasted quantity and the corresponding actual quantity.
- promotionA promotion is a planned commercial incentive or marketing activity intended to influence customer purchasing behavior.
- internal demand driverAn internal demand driver is a controllable factor affecting demand, such as pricing, promotions, or marketing.
- leading indicatorA leading indicator is a variable whose current or earlier movement provides information about a later outcome such as future demand.
- inventory policyAn inventory policy specifies how inventory is replenished and what inventory levels or targets are maintained under specified conditions.
- root mean squared errorRoot mean squared error is the square root of the mean of the squared differences between forecast values and corresponding actual values.
- ABC XYZ analysisABC XYZ classification is a two-dimensional inventory segmentation method that categorizes products by their relative value and demand variability or predictability.
- demand planning excellenceA standard of demand planning defined by effective decision support and efficient forecasting work.
- stockoutAn out-of-stock period is a time interval during which a product is unavailable because no usable inventory can be supplied to customers.
- bullwhip effectThe bullwhip effect is the amplification of demand, ordering, or inventory variability as it propagates through a supply chain.
- shortageA shortage is a condition in which available supply is insufficient to satisfy customer demand.
- apopheniaApophenia is the tendency to perceive meaningful patterns or connections among unrelated events or observations.
- confirmation biasConfirmation bias is the tendency to seek, notice, and favor information that supports existing beliefs while discounting or avoiding contradictory information.
- supply chain decisionA supply chain decision is a decision about the planning or execution of supply-chain activities such as ordering, production, or movement of goods.
- lost saleLost sales are customer purchases that would have occurred but were not completed because the requested product could not be fulfilled or another failure intervened.
- top-down forecastingTop-down forecasting predicts demand at an aggregate level and allocates or disaggregates that forecast to more detailed products or stock keeping units.
- groupthinkGroupthink is a pattern of group decision-making in which the desire for harmony and conformity suppresses independent evaluation and dissent.
- bottom-up forecastingBottom-up forecasting generates forecasts at detailed levels and aggregates them to higher levels.
- assumption-based discussionAn assumption-based discussion examines the underlying beliefs and demand drivers behind a forecast rather than focusing first on its final numerical value.
- scenario analysisScenario analysis is a technique for evaluating how alternative assumptions or input conditions affect predicted outcomes.
- intentional biasAn intentional bias is a systematic distortion introduced deliberately or strategically because incentives or objectives influence a forecasting judgment.
- flat splitA flat split is an allocation method that distributes an aggregate forecast across lower-level periods or entities using uniform proportions.
- overfittingOverfitting is a modeling condition in which a model fits historical data well but fails to generalize to future or otherwise unseen data.
- incentive mismatchIncentive misalignment is a condition in which participants' rewards or objectives encourage actions that conflict with the desired forecasting outcome.
- neural networkA neural network is a computational model composed of interconnected parameterized units organized in layers that transform inputs into outputs.
- inventoryInventory is the quantity of products held and available for sale or use.
- model selectionModel selection is the process of choosing a model structure or algorithm that is suitable for representing relationships in data and producing predictions.
- coefficient of variationThe coefficient of variation is the ratio of a quantity's standard deviation to its mean, often used to express relative variability.
- selective justificationSelective justification is an asymmetric requirement or acceptance of explanations for forecast adjustments or forecast errors.
- baselineA baseline model is the reference forecasting model against which forecast performance or later improvements are evaluated.
- multi-echelon inventory optimization modelMulti-echelon inventory optimization is the coordinated optimization of inventory decisions across multiple levels of a supply network.
- planning alignmentPlanning alignment is the coordination of supply, inventory, and related operational plans between organizations in a supply relationship.
- production and supply plansA supply plan specifies planned production, purchasing, and shipment quantities and coordinates the resulting logistics flow.
- wisdom of crowdsWisdom of crowds is the phenomenon in which aggregating diverse, independent judgments produces a more accurate collective estimate than most individual judgments.
- order batchingOrder batching is the practice of placing orders in infrequent, relatively large quantities instead of in smaller recurring quantities.
- decision treeA decision tree is a machine-learning model that makes predictions by applying a sequence of feature-based questions and following the resulting branches to a terminal prediction.
- shortage-censoringShortage-censoring is the exclusion or adjustment of observations from periods when recorded sales are limited by product shortages.
- judgmental forecastA judgmental forecast is a prediction of future demand that incorporates human assessment rather than relying solely on an automated forecasting model.
- mean absolute percentage errorMean absolute percentage error is the average across periods of absolute forecast error divided by the corresponding actual demand.
- short-term forecastShort-term forecasting predicts demand over a relatively near future horizon.
- global modelA global model is a forecasting model trained on data from an entire collection of products or series to learn patterns shared across them.
- local modelA local model is a forecasting model fitted independently to the historical data of one product or other individual series.
- feature engineeringFeature engineering is the selection, creation, or modification of input features for a machine learning model.
- one number forecastA one number forecast is a single forecast shared by stakeholders across a supply chain for coordinating their decisions.
- one number mindsetA one number mindset is an alignment approach in which stakeholders share assumptions, data, and a vision of the future without requiring identical forecasts.
- five-step demand planning frameworkThe five-step demand planning framework is a book-specific structured approach for improving demand planning and forecasting through five activities.
- Holt-Winters exponential smoothingExponential smoothing is a family of forecasting methods that updates estimates by weighting observations with exponentially declining importance as they become older.
- data cleaningData cleaning is the detection and correction of errors, inconsistencies, missing values, and unsuitable representations in datasets.
- statistical modelingA statistical model represents relationships in data using probabilistic or statistical assumptions to describe or predict observations.
- tree-based methodA tree-based machine learning model makes predictions using one or more decision-tree structures that represent relationships between inputs and outputs.
- accuracy targetA predefined level of forecast accuracy established as a performance objective.
- data collectionData collection is the systematic gathering of observations or records for analysis or modeling.
- forecastabilityForecastability is the degree to which a demand series can be predicted accurately from available historical information.
- supplier lead timeA supply lead time is the elapsed time between initiating a supply order and receiving the supplied goods.
- local averagingA moving average is a forecasting method that predicts a value by averaging observations from a recent fixed number of periods.
- inventory targetAn inventory target, or stock target, is the desired inventory quantity or level to maintain for a product over a planning period.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.