Forecasting and Sales & Operations Planning (S&OP)

Comprehensive vocabulary flashcards covering Forecasting and Sales & Operations Planning (S&OP) concepts, methodologies, mathematical models, and performance metrics from the lecture transcript.

Forecast

An estimate of the future level of some variable, most commonly a demand forecast of future sales used to determine capacity needs, business plans, and supply chain activities.

Demand Forecast

A forecast that estimates overall market demand or firm-level demand for products or services.

Supply Forecast

A forecast that provides information on the number of current producers and suppliers, projected aggregate supply levels, and technological or political trends.

Price Forecast

A forecast that predicts future price trends for key raw materials, inputs, and services.

First Law of Forecasting

The law stating that forecasts are almost always wrong, but they are still useful for operational planning within acceptable error limits.

Second Law of Forecasting

The law stating that forecasts for the near term tend to be more accurate due to greater data availability and fewer disruptive variables.

Third Law of Forecasting

The law stating that forecasts for groups of products or services tend to be more accurate because individual high and low errors balance each other out.

Fourth Law of Forecasting

The law stating that forecasts are no substitute for calculated values when firm, known demand is explicitly communicated.

Qualitative Forecasting Techniques

Forecasting methods based on intuition or informed expert opinion, typically applied when historical data is scarce, unavailable, or irrelevant.

Quantitative Forecasting Models

Forecasting models that use measurable or historical numerical data to generate predictions of future demand.

Market Surveys

A qualitative forecasting method involving structured questionnaires submitted to potential customers to gauge future demand.

Panel Consensus Forecasting

A qualitative forecasting method where a group of experts and key stakeholders come together in a joint meeting to develop a forecast.

Delphi Method

A qualitative forecasting technique where experts work individually using iterative anonymous questionnaires to converge on a consensus forecast.

Life Cycle Analogy Method

A qualitative forecasting method that estimates time frames and demand levels for a new product or service by drawing analogies to the growth trajectories of similar legacy products.

Build-Up Forecasts

A qualitative forecasting technique in which experts familiar with specific market segments estimate demand within their segments, which are then compiled into an aggregate forecast.

Time Series Forecasting Models

Quantitative forecasting models that arrange historical data in chronological order to observe patterns and project future demand.

Randomness

Unpredictable, unsystematic movement in time series demand from one period to the next without a recognizable pattern.

Trend

A persistent, long-term upward or downward movement observed in time series demand data over successive periods.

Seasonality

A repeated pattern of spikes or drops in a time series associated with specific recurring times of the year.

Cyclical Pattern

Demand variation driven by recurring long-term economic or industry changes occurring over multi-year horizons (such as $3$, $5$, or $7$ years).

Last-Period Model (Naïve Forecasting)

The simplest time series forecasting model, which sets the forecast for the upcoming period equal to the actual demand observed in the current period ($F_{t+1} = D_t$).

Simple Moving Average Model

A time series forecasting model that derives a forecast by calculating the unweighted arithmetic mean of actual demand over a specified number ($n$) of recent time periods.

Weighted Moving Average Model

A form of the moving average model that assigns specific percentage weights to past demand observations, where the sum of all weights must equal $1.00$ ($100\%$).

Exponential Smoothing Model

A time series forecasting model that calculates the next forecast as a weighted average of the current period's actual demand and current forecast using a smoothing constant parameter ($\alpha$).

Smoothing Constant ($\alpha$)

A parameter in exponential smoothing between $0$ and $1$; a lower value is used when demand is highly random to dampen noise, while a higher value is used when demand is stable to increase responsiveness.

Linear Regression

A statistical modeling technique that expresses a forecast dependent variable ($y$) as a linear function of an independent variable ($x$), in the form $y = a + bx$.

Causal Forecasting Models

A class of quantitative forecasting models in which the forecast variable is modeled as a function of one or more independent variables other than time.

Multiple Regression

A generalized form of linear regression that models a forecast dependent variable as a function of more than one independent variable.

Sales and Operations Planning (S&OP)

A process to develop long-term tactical plans by integrating marketing plans for products with supply chain management, also known as Aggregate Planning or Integrated Business Planning.

Strategic Planning

Planning that covers long time horizons (such as $2$ to $10$ years) with high risk, focusing on major capacity, facility, and capital investment decisions.

Tactical Planning

Intermediate-level planning (typically month-by-month over a $2$ to $18$ month horizon) that allocates workforce, inventory, subcontracting, and logistics resources under moderate risk.

Detailed Planning and Control

Short-term planning covering daily to weekly horizons with low risk, focusing on specific production scheduling, machine loading, and job assignments.

Top-Down Planning

An approach to S&OP where a single aggregated sales forecast drives the resource planning process, used when product mix and resource requirements are stable.

Bottom-Up Planning

An approach to S&OP used when the product/service mix is unstable and resource requirements vary significantly across offerings, requiring individual evaluation before aggregation.

Planning Values

Conversion factors used by decision-makers to translate financial sales forecasts into specific physical resource requirements, such as labor hours, equipment hours, or worker headcount.

Level Production Plan

An S&OP strategy in which production rate and workforce are held constant over time, using inventory or backorders to absorb differences between production and sales forecasts.

Chase Production Plan

An S&OP strategy in which production rates and workforce levels are varied in each period to match forecasted demand closely, minimizing inventory accumulation.

Mixed Production Plan

An S&OP strategy that varies both production rates and inventory levels in a customized combination to achieve the lowest overall operating cost.

Load Profile

A visual display of future capacity requirements based on released and planned orders over a specific planning horizon.

Net Cash Flow

The net flow of money into or out of a business over a given time period, calculated as total cash inflows minus total cash outflows.

Rolling Planning Horizon

A planning approach in which the S&OP plan is updated on a regular cycle (e.g., monthly), continuously extending the planning schedule forward by one period.

Yield Management

An approach used in service operations with perishable capacity, where prices are dynamically adjusted based on demand levels to maximize total profit.

Tiered Workforce

A workforce strategy that maintains a permanent core staff year-round and hires supplementary full-time, part-time, or seasonal employees during peak demand periods.

Offloading / Subcontracting

A strategy for smoothing workforce requirements by transferring work to outside suppliers or having customers perform part of the service process themselves.

Optimization Model

A mathematical model designed to find the best decision solution by maximizing or minimizing an objective function while adhering to operational constraints.

Objective Function

The target quantitative function in an optimization model that the decision-maker seeks to optimize (e.g., minimizing total operational cost).

Constraint

A quantifiable boundary condition or limitation imposed on an optimization model (e.g., maximum regular production capacity or allowable overtime).

Forecast Error ($FE_i$)

The numerical difference between actual observed demand and forecasted demand for a given period $i$, calculated as $FE_i = D_i - F_i$.

Mean Forecast Error (MFE)

The average of forecast errors over $n$ periods, used to evaluate whether a forecasting model has systematic bias (over-forecasting or under-forecasting).

Mean Absolute Deviation (MAD)

A measure of forecast accuracy calculated as the average of the absolute values of the forecast errors across all evaluated periods.

Mean Absolute Percentage Error (MAPE)

A measure of forecast accuracy that calculates the average absolute forecast error as a percentage of actual demand.

Tracking Signal

A measure used to monitor forecast performance over time to detect out-of-control forecast bias, typically expected to remain within limits of $-4$ to $+4$.