This unit covers the need for, different types of, and methods for forecasting demand. It is the first step in operations management and planning for changes in production and service capacity. Understanding statistical methods used in forecasting, and optimal levels of risk/uncertainty inherent in the analysis, is key to successful operations management.
Completing this unit should take you approximately 9 hours.
Demand management is a process for forecasting the anticipated demand for a product or service. This responsibility by the organization is highly planned, structured, and strategic so that the right amount of resources can be ordered, supplied, and managed appropriately. Moreover, data and information are vital components of forecasting as it provides objective reference points in order to predict future needs.
Watch this video. The speakers present forecasting methods based on overall business demand. How does a business reconcile the discrepancy between a forecast and actual demand?
Forecasting can be thought of as making predictions based on historical and current data to anticipate future needs. Quantitative forecasting is accomplished through objective numerical data and statistical analysis. In contrast, qualitative forecasting makes predictions using subjective knowledge guided by expertise or past experience.
This page gives a simple overview of both quantitative and qualitative methods. Study the forecasting diagram as it displays a visual representation of forecasting. When is it appropriate to use a qualitative forecast? A quantitative forecast?
Forecasting maintains a direct relationship with strategic planning. Predicting the future quantitatively helps a company manage its resources, navigate change, and mitigate negative market conditions. Generally speaking, this approach uses statistical confidence intervals and historical data to predict future trends.
Read this article where forecasting approaches, along with a hybrid forecasting method are covered. What types of data can a company use in quantitative forecasts?
Time series is best known as a collection of data points gathered at equally spaced points in time. It can be useful in various applications such as forecasting, budgeting, and quality control. When used in forecasting it can assist companies in predicting future values based on previously collected values.
Watch this video. Pay attention to the definition and explanation of time series. How would you define time series and what industries can use this approach?
Decomposition is a time-series task that deconstructs time-series data into categories. Through this deconstruction, the data can be broken down and analyzed in smaller portions to find patterns or variations. Essentially it provides a structured way of thinking through a time-series forecasting problem.
Read this article. It provides an overview of techniques associated with decomposition. Part 4, The Business Cycle, presents how this tool is applied in business operations. Why do you think decomposition is useful in understanding seasonality costs?
Linear regression strives to model a relationship between a dependent and independent variable. This analysis can be useful as a forecasting tool but also be used to quantify the strength of a relationship between two variables. Inferences can be made with the data because these relationships are not precise or perfect.
Read this chapter, which provides a general overview of regression. Focus on the Correlation and covariance section. How would you define correlation and covariance?
A moving average is also known as a rolling average or running average to analyze data points. This statistical approach is commonly used with time series data to identify trends. When used in conjunction with time series it can filter out high-frequency components and smooth out data.
Watch this video. Consider the different types of moving averages that are presented. Can you clearly articulate the differences in your own words?
Exponential smoothing is a type of moving average and refers to a technique that is applied to time-series data for forecasting. When applied, it allows one to estimate the demand for items accounting for intermittent and seasonal variations. It is relatively simple and is widely used and accepted for short-term forecasting.
Watch this video, and note how the data is manipulated and forecasted. Can you identify any patterns?
Supply and demand are key concepts fundamental to economic analyses. The demand part of this relationship plays a huge role in the price of a good or service. While there are many elements of demand, all generate a force on the supply chain requiring each to adjust accordingly.
Read this article. The authors suggest that underestimating demand for electricity services can result in shortages that impact productivity and overall economic growth. Box 1 presents a case study of electricity forecasting in Armenia. How do you think your personal energy consumption helps or hurts your country's economic development?
Trends in Demand indicate medium and long-term fluctuations when the desire for a product or service increases or decreases over time. All products and services cycle through peaks as well as drops in demand. This could happen due to factors that may include word of mouth, marketing campaigns, changes in demographics, or shifts in interest.
Watch this video. It explains forecasting for the short, medium, and long term. What trends can you identify in the airline sector and how do you think these trends affect the price you pay?
Seasonality is another component of demand which can include certain weeks, months, years, or any time period. Since this area is irregular and sporadic, it is further proof that seasonal forecasting must be as accurate as possible. Both historical data and real-time data are critical to anticipating this cyclical dynamic.
Read this article. It is important because seasonal demand is addressed as the authors attempt to successfully predict demand. In your experience, what are some seasonal products or services that you purchase?
A causal relationship attempts to identify a connection between two specified factors. This quantitative forecasting tool suggests that one factor causes changes in the other factor. While the first event is known as the cause, the following event is known as the event.
Read this article. It posits that investments in information technology enhance supply chain business performance. As you read the Literature Review, think about how advances in technology have increased your own productivity.
When it comes to statistics, forecasting error is defined as the difference between the real data value and the predicted data value. The error itself is calculated as the outcome minus the value of the original forecast. The greater the error in forecasting the greater potential to compromise economic and performance outcomes.
Watch this video. Pay particular attention to the presentation of a summary of forecasting errors. In your personal life, what happens when your forecast of costs for a vacation holiday is less than your actual costs?
Qualitative forecasting is unique in that it uses experiential, anecdotal, or other non-numerical means to predict demand. Forces that affect this can be attributed to weather, the economy, world events, or even catastrophes. This can be useful in large-scale resource planning and also conflict resolution.
Read this article. Two forecasting approaches are employed for forest fire disaster response planning. Focus on the qualitative flow chart in Figure 2.
This review video is an excellent way to review what you've learned so far and is presented by one of the professors who created the course.
Watch this as you work through the unit and prepare to take the final exam.
We also recommend that you review this Study Guide before taking the Unit 5 Assessment.
Take this assessment to see how well you understood this unit.