Supply Chain: Demand Planning (Sales Forecasting and S&OP)

Explore regression analysis as a core time series method for forecasting sales by modeling the relationship between a dependent variable and independent variables such as time and marketing activity.

Apply regression analysis to forecast sales by computing the intercept and slope from two years of data, then use y = a + b x and evaluate accuracy with MAPE.

Exponential smoothing uses historical data to reveal long-term seasonality by smoothing fluctuations and weighting relevant data, producing forecasts with F_{t+1}=αD_t+(1−α)F_t, ideal for seasonal data.

Learn exponential smoothing to forecast sales by weighting actual demand and prior forecast with a smoothing constant alpha, starting values, and a best alpha chosen via MAPE and goal seek.

A sales forecast projects the expected bookings at the end of a measurement period, typically weekly to quarterly, using qualitative and quantitative techniques, including top-down or bottom-up approaches.

Compare qualitative and quantitative forecasting techniques, showing when to use each: qualitative for products or markets with no historical data, using jury of executive opinion; quantitative relies on mathematical equations.

Compare top-down and bottom-up forecasting: macro-driven board decisions versus micro, territory-based sales data, considering population growth, customer preferences, and supply capabilities to build SKU-level forecasts.

Learn to create a balanced forecast by integrating top-down and bottom-up figures with qualitative and quantitative techniques to ensure accuracy, open communication, and attainable figures.

Balance forecasting approaches to improve sales forecasts and spot early warnings for the business. Blend historical data, linear regression, and qualitative inputs across regions for items, launches, and market events.

Integrate supply and demand planning by aligning intersection points, shared building blocks, and overlap areas. Improve unconstrained demand handling, forecast accuracy, and inventory control through cross-functional collaboration.

Adopt a rolling forecast to replace static budgets, maintaining a constant horizon of five quarters or 15 months through add drop updates, enabling driver-based, project-based planning for timely decisions.

Explore rolling forecasts and the C to D ratio, analyze quarterly variances, and assess supply readiness to ensure demand can be met under flat or changing demand scenarios.

Use rolling forecasts and C to D ratio to analyze seasonal demand, comparing cycle horizon capacity to demand and identifying deficits for timely capacity investments.

Time fence policy defines when changes to the master production schedule are allowed by splitting forecasts into frozen, slushy, and liquid zones, to minimize costs and protect existing sales orders.

Apply the time fence policy to demand planning by assessing frozen, slushy, and liquid zones across forecast versions. Ensure S&OP balance and avoid costly variances.

Analyze fixed fiscal year sales forecast variance alongside monthly actuals versus budget to guide timely adjustments and ensure the annual target is met.

Learn to measure sales forecast accuracy using simple methods and MAPE, set tolerance ranges, and assess category and item accuracy to align forecasts with actual sales.

Explore the debate over MAPE denominator, compare using forecast versus actual, and learn why the actual should be the denominator to avoid manipulation and inventory risks.

Improve forecast accuracy for sales forecasts by enabling an enterprise wide data flow, reducing inventory and carrying costs, and strengthening cash flow and revenue through better demand planning.

Explore a collaborative demand planning and S&OP process to improve forecast accuracy through regular reviews, a unified forecast model, Pareto analysis, and actions on promotions, pricing, inventory, and supply risks.

Show that focusing on where they missed the forecast is not the place they should look.