Forecasting Process and Organizational Integration

Introduction

Demand forecasting is not merely the production of a numerical estimate. It is an organizational process through which evidence about future demand is collected, interpreted, challenged, agreed, communicated, and converted into decisions about procurement, production, inventory, transport, capacity, finance, and customer service. A technically accurate forecast can still fail if it is produced too late, owned by no one, distorted by incentives, or disconnected from planning decisions.

In logistics, the forecasting process must therefore be designed as an end-to-end management cycle. The cycle begins with data preparation and statistical forecasting, but it extends through segmentation, exception management, commercial review, consensus formation, executive approval, execution, and measurement. Modern organizations also integrate forecasting with Sales and Operations Planning (S&OP), Integrated Business Planning (IBP), and Collaborative Planning, Forecasting and Replenishment (CPFR).

The central managerial question is not simply, “What will demand be?” It is also, “Who should decide, using which evidence, at what level of aggregation, within what time horizon, and with what consequences?”

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Design an end-to-end organizational forecasting cycle.
2. Allocate forecasting ownership and accountability across functions.
3. Segment demand according to business relevance, variability, value, and predictability.
4. Design exception-management rules that focus attention on material risks.
5. Explain the role of statistical, commercial, consensus, and collaborative forecasts.
6. Integrate forecasting with S&OP, IBP, and CPFR.
7. Evaluate forecasting performance using accuracy, bias, value-added, and process measures.
8. Identify governance, ethical, technological, and implementation risks.

1. Forecasting as a Cross-Functional Process

A forecast is an estimate of future demand made at a specified time, for a specified product or service, location, customer group, and time bucket. Forecasts are conditional statements: they depend on assumptions about price, promotions, economic conditions, availability, lead times, and customer behavior.

A useful distinction is between three related quantities:

• Demand forecast: the expected customer requirement, whether or not the organization can supply it.
• Sales forecast: the expected quantity recorded as sales, which may be constrained by stock-outs or capacity.
• Operational plan: the feasible response involving inventory, production, purchasing, transport, and capacity.

Confusing these quantities creates serious errors. If a stock-out reduces recorded sales, using sales as demand without adjustment may cause the next forecast to fall even though underlying demand remains strong. Conversely, a sales target may be deliberately aspirational and should not automatically be treated as an unbiased demand estimate.

Forecasting has four organizational characteristics:

1. It is hierarchical. Forecasts may be required by SKU, location, customer, channel, product family, region, and total company.
2. It is time-dependent. Strategic, tactical, and operational horizons require different data, methods, and decisions.
3. It is collaborative. Relevant knowledge is distributed among marketing, sales, supply, finance, customers, and suppliers.
4. It is decision-oriented. Forecasts matter because they influence resource allocation and service outcomes.

A sound process separates the creation of evidence from the authorization of decisions. Analysts should be able to produce a transparent statistical baseline; commercial teams should add documented information about market events; supply planners should test feasibility; and executives should resolve trade-offs rather than arbitrarily overwrite numbers.

2. The End-to-End Forecasting Cycle

A mature forecasting cycle is closed-loop rather than one-directional. It learns from forecast errors, changes in market conditions, and the consequences of decisions.

Stage 1: Define the forecasting purpose

The organization first specifies the decision the forecast will support. A forecast for ocean-freight capacity may require a monthly regional view six months ahead, whereas a warehouse replenishment forecast may require daily SKU-location estimates over several weeks. The purpose determines the horizon, granularity, frequency, accuracy target, and owner.

The forecast specification should state:

• forecast object and hierarchy;
• time bucket and horizon;
• demand measure and units;
• inclusion or exclusion of promotions, returns, cancellations, and stock-outs;
• responsible owner and review forum;
• decisions affected by the forecast.

Stage 2: Prepare and govern data

Data preparation includes cleansing duplicate transactions, aligning calendars, correcting unit-of-measure inconsistencies, identifying product substitutions, and distinguishing genuine demand changes from supply constraints. Master-data governance is particularly important when products are introduced, discontinued, repackaged, or moved between distribution centres.

A demand history should record relevant causal events such as price changes, promotions, competitor actions, weather disruptions, public holidays, and lost sales. It should also preserve an audit trail showing when data was changed and by whom.

Stage 3: Generate the statistical baseline

The statistical baseline provides a reproducible starting point. Depending on the demand pattern, methods may include naïve forecasts, moving averages, exponential smoothing, regression, intermittent-demand methods, or machine-learning models. The baseline should be evaluated against simple benchmarks. A complex algorithm is not justified merely because it is complex.

For a one-period naïve forecast:

where F_t+1 is the forecast for period t+1 and A_t is actual demand in period t. The method assumes that the most recent observation is a useful estimate of the next period and is often an important benchmark.

A moving-average forecast using the previous k observations is:

Here, k is the number of observations, and A_t-i is demand in an earlier period. This method is appropriate when recent observations are reasonably representative and no strong trend or seasonality is present. Larger k values smooth noise but react more slowly to change.

Stage 4: Segment the portfolio

Forecasting resources should not be distributed equally across all items. Segmentation determines which items receive detailed modelling, human review, collaboration, or simple rules.

Common segmentation dimensions include:

• Value: annual consumption value, revenue, margin, or strategic importance.
• Demand variability: coefficient of variation, intermittency, or forecast-error volatility.
• Predictability: stability of patterns and strength of causal drivers.
• Criticality: consequences of shortage, including safety, contractual, or production impact.
• Lifecycle: introduction, growth, maturity, decline, and discontinuation.
• Customer or channel: strategic accounts, e-commerce, distributors, or internal users.

The coefficient of variation is often used for relative variability:

where σ_D is the standard deviation of demand and μ_D is mean demand over the selected period. Both quantities must use the same units, so CV is dimensionless. The measure is meaningful when the mean is positive and sufficiently stable; it becomes unstable when the mean is close to zero.

An ABC-XYZ matrix illustrates how value and predictability can be combined. An AX item has high value and relatively stable demand and may justify close monitoring and sophisticated replenishment. A CZ item has low value and highly variable demand and may be managed using simple rules, order-on-demand policies, or broader aggregation. The matrix is a prioritization device, not a universal law: a low-value item may still be operationally critical.

Stage 5: Conduct exception management

Exception management prevents meetings from becoming reviews of every forecast line. The system identifies material deviations from expected behaviour and routes them to accountable owners.

Possible exception triggers include:

• forecast change beyond a percentage or absolute threshold;
• demand outside a statistically expected range;
• large forecast error or persistent bias;
• promotion not reflected in the baseline;
• capacity or supplier constraint;
• inventory projected below a service threshold;
• new product with inadequate history;
• forecast override exceeding a governance limit.

Thresholds should be differentiated by segment. A 20% change on a high-volume, high-margin product may be more important than a 100% change on a low-volume item. Alerts should therefore combine magnitude, value, probability, and business impact.

Exception management has two failure modes. Alert overload occurs when too many items are flagged and planners ignore the system. Blind spots occur when thresholds are too narrow or data quality is poor. Thresholds should be reviewed using historical performance and user feedback.

Stage 6: Add structured business intelligence

Sales, marketing, operations, and customers may possess information not visible in historical demand. Examples include a confirmed tender, a competitor withdrawal, a new contract, a price increase, a campaign, or a distribution-centre closure.

Business input should be structured rather than informal. Each adjustment should identify:

• the quantity and periods affected;
• the causal event;
• the confidence level;
• supporting evidence;
• whether the effect is incremental, temporary, or a shift in the baseline;
• the responsible person and expiry date.

This approach distinguishes genuine information from optimism, target setting, or political negotiation.

Stage 7: Create a consensus forecast

The consensus forecast reconciles the statistical baseline, commercial intelligence, customer collaboration, supply constraints, and financial assumptions. Consensus does not mean averaging every opinion. It means reaching an explicit, documented view of unconstrained demand and, where necessary, a separate feasible supply plan.

A simple weighted combination may be used when forecasts are produced by methods with known and relatively stable performance:

subject to:

where F_c is the combined forecast, F_i is forecast i, and w_i is its weight. The weights must sum to one so that the combined forecast remains in the same units. Weights should be based on validated performance or a clearly documented decision rule; they should not simply reflect organizational seniority.

The consensus process should preserve at least three views where relevant: baseline demand, adjusted unconstrained demand, and constrained or executable demand. Collapsing these into one number hides trade-offs and encourages supply limitations to contaminate demand estimates.

Stage 8: Reconcile and approve

The forecast is reconciled across product, location, customer, and time hierarchies. Reconciliation seeks to prevent contradictory numbers—for example, regional forecasts that do not add to the corporate forecast. It also exposes assumptions about market growth, capacity, working capital, and service.

Approval should occur in a defined governance forum. Approval is not a guarantee that the forecast will be correct; it is authorization to use a stated planning assumption. Material risks and unresolved disagreements should be recorded rather than concealed.

Stage 9: Publish, execute, and monitor

The approved forecast is released to the systems and teams that use it. Changes after the freeze period should be controlled because frequent uncontrolled changes destabilize procurement, production, and transport.

Actual demand and outcomes are then monitored. The cycle begins again with error analysis, root-cause review, and method or process improvement.

3. Forecast Ownership and Governance

Ownership means accountability for process quality and decision readiness; it does not mean that one person must perform every forecasting task. A practical governance model contains the following roles:

The organization should define decision rights using a responsibility matrix. For example, the demand planner may be responsible for publishing the statistical baseline, sales may be consulted on a major customer event, finance may validate monetary implications, and the executive forum may approve a capacity trade-off.

Governance should include version control, override logs, approval deadlines, model documentation, data-quality checks, and escalation procedures. A forecast override is not inherently bad. It is valuable when it adds information unavailable to the model and harmful when it merely reflects incentives or unsupported optimism.

4. Measuring Forecasting and Process Performance

No single metric captures forecast quality. Measures should be selected according to the decision context and interpreted with knowledge of demand intermittency and scale.

For actual demand A_t and forecast F_t over n periods, the error may be defined as:

Positive error means actual demand exceeded the forecast. Mean error, often called mean forecast error, is:

ME indicates directional bias but can conceal offsetting errors.

Mean absolute error is:

MAE is expressed in demand units and is useful for operational interpretation. Root mean squared error is:

RMSE penalizes large errors more heavily than MAE and is useful when large misses are disproportionately costly.

Mean absolute percentage error is:

MAPE is undefined when A_t = 0 and can become unstable when actual demand is very small. It should therefore not be used uncritically for intermittent demand. Weighted or scaled measures may be more suitable for portfolios with many zeros.

Forecast bias can also be monitored using cumulative error:

A positive CFE indicates cumulative under-forecasting under the stated error convention. Bias should be interpreted alongside service, inventory, and commercial context: deliberate protection against severe shortages may produce a positive forecast bias but still be economically rational.

Process measures are equally important. They include forecast cycle adherence, percentage of exceptions resolved on time, override frequency, data-quality defects, forecast-value-added results, stakeholder participation, and the timeliness of publishing. A forecast process that is accurate but consistently late may be operationally inferior to a slightly less accurate process that supports decisions on time.

5. Integration with S&OP and IBP

Sales and Operations Planning

S&OP is a recurring cross-functional process that aligns demand, supply, inventory, capacity, and financial implications over a medium-term horizon. A typical monthly cycle includes:

1. data gathering and portfolio review;
2. demand review;
3. supply review;
4. pre-S&OP reconciliation;
5. executive S&OP decision meeting;
6. publication and follow-up.

The demand review does not exist to ratify a sales target. It examines the baseline, assumptions, risks, promotions, new products, lost sales, and customer intelligence. The supply review tests whether the demand plan can be delivered within capacity, labour, inventory, supplier, and transport constraints.

S&OP is particularly useful when trade-offs are cross-functional. For example, an organization may choose overtime production, inventory pre-build, allocation to priority customers, or a revised promotion. The forecast supplies an assumption; S&OP converts that assumption into an integrated plan.

Integrated Business Planning

IBP extends S&OP by connecting operational planning more explicitly to strategy, finance, portfolio decisions, and long-range capacity. It may include scenarios for market entry, network redesign, capital investment, product rationalization, and strategic customer commitments.

The distinction between S&OP and IBP is not universally standardized. In some organizations, IBP is a more mature form of S&OP; in others, it is a broader governance label. The essential principle is vertical and horizontal integration: strategic goals should be translated into feasible operational plans, and operational constraints should inform strategic decisions.

Forecasts used in IBP should be scenario-based where uncertainty is material. A single-point estimate may be supplemented by optimistic, central, and adverse cases. Scenario planning is not a replacement for a most-likely forecast; it reveals the consequences of uncertainty and supports contingency design.

6. CPFR and Interorganizational Collaboration

Collaborative Planning, Forecasting and Replenishment extends forecasting beyond organizational boundaries. Suppliers, manufacturers, distributors, and retailers share selected information and coordinate planning activities.

A CPFR process commonly includes:

• establishing a collaborative relationship and business objectives;
• agreeing on roles, data, calendars, and performance measures;
• creating a joint business plan;
• developing forecasts independently or jointly;
• identifying forecast exceptions;
• resolving exceptions through dialogue;
• translating the agreed forecast into replenishment orders;
• measuring results and improving the relationship.

CPFR can reduce the bullwhip effect when partners share reliable sales, inventory, promotion, and shipment information rather than reacting only to orders. However, collaboration does not automatically improve results. It requires compatible data definitions, trust, incentives, confidentiality safeguards, and agreed rules for handling disagreement.

The partners should distinguish point-of-sale demand, orders, shipments, and inventory. Sharing only order history may preserve distortions caused by batch ordering, allocation, or shortages. Collaborative forecasts should also define ownership of promotional assumptions and the process for revising them.

7. Worked Application: Distribution Network Forecast Cycle

A regional distributor supplies 1,000 cases per month of a mature product. The most recent three months of observed demand are 920, 1,000, and 1,080 cases. A three-month moving-average baseline for the next month is:

The sales team provides evidence of a confirmed promotion expected to add 180 cases. The demand planner records this as an incremental event rather than replacing the baseline. The adjusted unconstrained demand forecast is therefore:

The supply planner identifies that available supplier capacity supports only 1,100 cases during the period. The organization should not revise the demand forecast downward to 1,100. Instead, it should retain:

• baseline demand: 1,000 cases;
• promotional increment: 180 cases;
• unconstrained demand: 1,180 cases;
• feasible supply: 1,100 cases;
• potential shortage: 80 cases.

The S&OP forum may choose among overtime, partial allocation, alternative sourcing, customer communication, or acceptance of the shortage. Each choice has different service, cost, and relationship consequences. After the month closes, actual demand, lost sales, fill rate, and promotion effectiveness are reviewed. If actual recorded sales were 1,100 because of the supply limit, analysts must investigate whether 80 cases of demand were lost before judging forecast accuracy.

This example illustrates why demand forecasting and supply planning should be connected but not confused.

8. Illustrative Case: Northstar Home Delivery

Background

Northstar Home Delivery is a fictional national distributor of household appliances. It operates three distribution centres and supplies retailers, online channels, and service contractors. Forecasts are generated by region and product family, but each function maintains separate spreadsheets.

Problem

The company experiences excess inventory in slow-moving appliances while frequently expediting fast-moving spare parts. Sales representatives routinely increase forecasts before quarterly target discussions. Supply planners reduce forecasts when suppliers report shortages. No one can explain which version of the forecast was used for purchasing. Forecast accuracy appears acceptable at total-company level but is poor at SKU-location level.

Analysis

A diagnostic review identifies five causes:

1. sales history is not adjusted for stock-outs;
2. demand and supply forecasts are combined into one number;
3. high-value and intermittent items receive the same review frequency;
4. overrides are undocumented;
5. S&OP meetings focus on explaining last month rather than deciding future actions.

The company also uses MAPE for items with many zero-demand periods, producing unstable and misleading percentages.

Approach and solution

Northstar designs a twelve-week implementation:

• It establishes a common demand dictionary distinguishing orders, shipments, sales, and estimated lost sales.
• It introduces an automated statistical baseline with naïve and seasonal benchmarks.
• It classifies items using value, variability, criticality, and lifecycle.
• It creates exception rules for major value changes, persistent bias, promotions, and service risk.
• It requires every override to include a reason, quantity, period, evidence, owner, and expiry date.
• It establishes separate unconstrained demand and feasible supply views.
• It redesigns S&OP into data, demand, supply, reconciliation, and executive decision stages.
• It begins CPFR with two strategic retailers using point-of-sale and inventory data.
• It replaces MAPE for intermittent spare parts with unit-based and scaled error measures.

Outcome

After two planning cycles, the company has not yet eliminated forecast error, but it reduces meeting time and improves decision transparency. Planners spend more time on high-impact exceptions, procurement can distinguish market growth from supplier limitations, and executives can evaluate the cost of alternative capacity decisions. Longer-term performance improvement is assessed using inventory, service, expedites, forecast bias, and process adherence rather than one accuracy percentage.

Learning points

The case demonstrates that forecasting transformation is principally a process and governance intervention. Technology improves access and speed, but it cannot resolve unclear definitions, conflicting incentives, or absent decision rights. The separation of demand, supply, and financial views is essential for meaningful cross-functional dialogue.

9. Technology, Analytics, and Artificial Intelligence

Forecasting platforms can automate data ingestion, hierarchy management, model selection, scenario analysis, workflow, and exception alerts. Machine-learning methods may identify nonlinear relationships involving price, promotions, weather, digital traffic, or macroeconomic indicators. They are most useful when relevant explanatory data is available, stable enough for learning, and governed for quality.

Artificial intelligence should be used with human and organizational controls. Important controls include:

• documented training data and feature definitions;
• prevention of data leakage from future information;
• out-of-sample validation and benchmark comparison;
• monitoring for drift after market or policy changes;
• explainable drivers for material adjustments;
• access controls and protection of commercially sensitive data;
• human review for exceptional, high-impact decisions.

Automation should reduce low-value manual work, not remove accountability. Generative systems may summarize exceptions or propose explanations, but they should not silently change approved forecasts or invent evidence. Model performance should be evaluated against business outcomes and decision costs, not only statistical fit.

10. Challenges, Ethics, and Limitations

Forecasting is affected by structural uncertainty that no model can eliminate. New products lack history; markets experience regime changes; promotions alter timing; supply shortages censor observed demand; and competitor actions may be unobservable.

Important organizational risks include:

• incentive bias: teams manipulate forecasts to protect targets or capacity;
• aggregation bias: accurate totals conceal poor local decisions;
• automation bias: users accept a model because it appears objective;
• data privacy: customer-level data may reveal sensitive commercial behaviour;
• algorithmic discrimination: allocation or service decisions may systematically disadvantage customer groups;
• confidentiality risk: CPFR data may expose strategic information;
• false precision: point forecasts may be interpreted as certainty.

Ethical governance requires transparency about assumptions, proportional access to data, documented overrides, appropriate retention and security, and escalation when a forecast affects critical goods or vulnerable customers. Managers should consider prediction intervals or scenario ranges when decisions are sensitive to uncertainty. A forecast should communicate what is known, what is assumed, and what could change.

11. Best-Practice Design Principles

A robust forecasting process follows these principles:

1. Begin with the decision, not the algorithm.
2. Define demand consistently across functions and partners.
3. Preserve a transparent statistical baseline.
4. Segment effort according to value, variability, criticality, and predictability.
5. Use exception management to focus scarce managerial attention.
6. Require evidence and expiry dates for overrides.
7. Separate unconstrained demand from supply feasibility.
8. Integrate volume, value, capacity, inventory, and service in S&OP or IBP.
9. Collaborate externally only with clear data definitions and governance.
10. Measure bias, error, process quality, and business outcomes together.
11. Compare advanced models with simple benchmarks.
12. Treat forecast improvement as a continuous learning cycle.

Chapter Summary

Forecasting is an end-to-end organizational cycle rather than an isolated statistical exercise. It begins by defining the decision and forecast specification, continues through data governance, statistical baseline generation, segmentation, exception management, structured business input, consensus formation, reconciliation, approval, execution, and performance learning.

Ownership must be distributed across demand planning, sales, marketing, supply, finance, analytics, executives, and external partners. Governance establishes decision rights, version control, override rules, and accountability. Segmentation using value, variability, predictability, criticality, and lifecycle enables differentiated effort. Exception management focuses attention on material risks while avoiding alert overload.

Forecast accuracy should be assessed using suitable measures such as ME, MAE, RMSE, and carefully applied percentage or scaled measures. Bias and process measures are essential complements. S&OP links demand and supply decisions; IBP extends the connection to strategy and finance; CPFR coordinates forecasting and replenishment across organizational boundaries.

The most effective forecasting systems combine statistical evidence, human knowledge, collaborative information, transparent assumptions, and disciplined decision governance. Technology and AI can improve speed and analytical reach, but they cannot substitute for clear definitions, trustworthy data, appropriate incentives, and accountable management.

Key Terms

Forecast: An estimate of future demand for a defined object, horizon, and time bucket.

Demand forecast: Expected customer requirement, independent of supply constraints.

Sales forecast: Expected recorded sales, which may be affected by availability.

Statistical baseline: A reproducible forecast generated from historical or explanatory data before business adjustments.

Forecast horizon: The future period covered by a forecast.

Forecast hierarchy: Linked levels such as SKU, location, product family, region, and total business.

Segmentation: Division of items or customers into groups requiring different forecasting policies.

Coefficient of variation: Standard deviation divided by mean demand, measuring relative variability.

Exception management: Identification and routing of material deviations for investigation.

Override: A documented change to a system-generated or previously approved forecast.

Consensus forecast: An agreed forecast integrating statistical and cross-functional evidence.

Unconstrained demand: Demand expected if product and capacity limitations do not restrict supply.

S&OP: Cross-functional process aligning demand, supply, inventory, capacity, and finance.

IBP: An extended planning process connecting operational plans with strategy and financial decisions.

CPFR: Collaborative Planning, Forecasting and Replenishment among supply-chain partners.

Forecast bias: Persistent tendency to over-forecast or under-forecast.

MAE: Mean absolute error, the average absolute forecast error in demand units.

RMSE: Root mean squared error, a measure that gives greater weight to large errors.

Forecast value added: Evaluation of whether a process step improves the forecast relative to its predecessor.

Bullwhip effect: Amplification of demand variability as information moves upstream in a supply chain.

Review Questions

Short Answer Questions

1. Distinguish a demand forecast from a feasible supply plan.
2. What is the purpose of a statistical baseline?
3. Why should forecast overrides be documented?
4. State two limitations of MAPE.
5. What is the role of exception management?

Long Answer Questions

1. Explain the stages of an end-to-end forecasting cycle and show how they connect to logistics decisions.
2. Discuss how value, variability, criticality, and lifecycle can be used to segment a forecasting portfolio.
3. Compare the roles of demand planning, sales, supply planning, finance, and executives in forecast governance.
4. Explain how S&OP and IBP differ in scope and how forecasting supports each process.
5. Discuss the conditions required for CPFR to improve supply-chain performance.

Analytical / Application Questions

1. A product has demand observations of 900, 1,050, and 1,200 units. Calculate the three-period moving-average forecast and explain one limitation of this method.
2. A promotion adds 250 units to a baseline forecast of 1,400 units, but supply capacity is limited to 1,500 units. Present the baseline, unconstrained demand, feasible supply, and implied shortage, and explain how the issue should be handled in S&OP.
3. Design an exception-management policy for a portfolio containing high-value stable items, low-value intermittent items, new products, and critical service parts. Justify different review rules for each segment.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes a demand forecast from a supply plan?

A. A demand forecast always equals the production target.

B. A demand forecast estimates customer requirements, while a supply plan represents a feasible response.

C. A demand forecast includes only quantities that suppliers can deliver.

D. A supply plan is generated without considering capacity.

Correct Answer: B. A demand forecast estimates customer requirements, while a supply plan represents a feasible response.

Question 2

What is the main purpose of a statistical baseline?

A. To provide a transparent and reproducible starting forecast

B. To replace all commercial judgement

C. To guarantee that actual demand will be met

D. To convert demand into financial profit automatically

Correct Answer: A. To provide a transparent and reproducible starting forecast

Question 3

If actual demand is 1,200 units and the forecast is 1,000 units, using e_t = A_t − F_t, what is the forecast error?

A. −200 units

B. 0 units

C. 200 units

D. 1,000 units

Correct Answer: C. 200 units

Question 4

Which measure is expressed in the same units as demand?

A. Coefficient of variation

B. MAE

C. MAPE

D. Forecast percentage bias only

Correct Answer: B. MAE

Question 5

Why can MAPE be unsuitable for intermittent demand?

A. It cannot be calculated when actual demand is zero.

B. It always produces negative values.

C. It is measured in kilograms only.

D. It ignores the forecast value completely.

Correct Answer: A. It cannot be calculated when actual demand is zero.

Question 6

Which activity is characteristic of exception management?

A. Reviewing every item with equal frequency

B. Ignoring large changes until the annual budget cycle

C. Routing material deviations to accountable owners

D. Replacing all forecasts with sales targets

Correct Answer: C. Routing material deviations to accountable owners

Question 7

What is the principal purpose of S&OP?

A. To align demand, supply, inventory, capacity, and financial implications

B. To calculate only warehouse picking productivity

C. To eliminate the need for forecasts

D. To determine employee compensation

Correct Answer: A. To align demand, supply, inventory, capacity, and financial implications

Question 8

What does CPFR primarily involve?

A. Internal budgeting without supply-chain partners

B. Collaborative planning, forecasting, and replenishment across organizations

C. Replacing point-of-sale data with purchase orders

D. Allowing suppliers to set customer prices

Correct Answer: B. Collaborative planning, forecasting, and replenishment across organizations

Question 9

A baseline forecast is 1,000 units and a documented promotion adds 180 units. What is unconstrained demand?

A. 820 units

B. 1,000 units

C. 1,180 units

D. 1,800 units

Correct Answer: C. 1,180 units

Question 10

Which practice most improves forecast-governance transparency?

A. Allowing undocumented manual changes

B. Combining demand and supply constraints into one unexplained number

C. Maintaining override reasons, evidence, owners, and expiry dates

D. Using only executive judgement

Correct Answer: C. Maintaining override reasons, evidence, owners, and expiry dates