Overview
Please review all three Articles, and provide a general review of the students response to the article. Three Articles: 100 - 150 words per article.
Article One
Forecasting can be classified into four basic types: qualitative, time series analysis, casual relationships and simulation.
Time Series analysis, is a data driven idea of the past demands that can predict the future. An example for this is survey sheets from customers. Just like the surveys we get from stores or vendors that we just purchased something from. That is how they are predicting what are the future demands for their store.
Casual forecasting, assumes that demand is related to some underlying factor or factors in the environment. This is a fun one, this is depending on the weather and what are the “in” things happening during that time. Example during the summertime, ice cream parlors are on demand because it’s hot and ice cream cools people off.
Simulation models allow the forecaster to run through a range of assumptions about the condition of the forecast. This is a model where a company will predict their future sales of a new product. For example, when Apple releases a new Apple watch, the company starts a simulation to predict what their future sales are.
Qualitative forecasting is method making predictions based on historical, expert opinions or subjective judgement. This is a method used when historical data is limited, and outdated.
References:
Jacobs, F. R., & Chase, R. B. (2024). Operations and supply chain management (17th ed.)
Article Two
Forecasting helps businesses predict what might happen in the future based on past data or expert opinions. There are two main types, qualitative and quantitative methods. Qualitative methods are best when there’s little or no data. For example, the Delphi method gathers expert opinions to make predictions, like estimating demand for a brand-new tech gadget. Market research uses surveys or focus groups to learn what customers want which is helpful when launching new products (Rowe et al., 1999). Sales force composite relies on sales teams to predict future sales based on what they hear from customers, which works well for local or regional forecasts.
Quantitative methods, on the other hand, use numbers and stats. The naive method assumes the next period will look just like the last, simple but works when things are stable. A moving average smooths out ups and downs by averaging past data, and exponential smoothing does the same but gives more weight to recent trends which is great for short-term forecasting. If sales change with the seasons, seasonal indexes help adjust predictions accordingly. For more complex forecasts, regression analysis looks at how different factors, like advertising or interest rates, affect outcomes. Econometric models take it a step further by using economic theory and stats to predict big-picture trends like GDP growth. Each method has its own best use depending on the situation and available data.
Reference
Rowe, G., Wright, G., & Bolger, F. (1999). The Delphi technique as a forecasting tool: Issues and analysis. International Journal of Forecasting, 15(4), 353–375. https://doi.org/10.1016/S0169-2070(99)00018-7
This is a great discussion question. If I worked for a company that is preparing to launch a new breakfast cereal and I was asked to propose forecasting demand for it, here is how I would use both a qualitative and quantitative method. For a qualitative technique, I feel that a market survey would be beneficial to understand if a minimal preparation option is an actual preference by our consumers. As proposed in the referenced article, a market survey could provide valuable data through an online survey that could reach many people in a specific demographic for which we are aiming, and do things such as predict market dynamics and consumer behavior (Caniato, 2011). Once the survey is completed, we could use historical data to understand if there ever was a demand for low preparation options, and use our quantitative technique. In this case, has a company ever developed such a product in the past? We could understand if there ever was a demand, if the product failed, and using a regression analysis we could explore the relationship between the demographics back then and what they currently are. for example we could understand if there was a need for a minimal preparation cereal before so much of our population worked from home. Now, since more people work from home, maybe they are ok with a cereal that requires some more preparation than previously demanded. These techniques compliment each other by allowing us to see and understand both types of data. An analysis that takes both statistical and opinion based data for sales of a cereal should not be rocket science, and I dont think we should reinvent the wheel in this case.
Reference:
- Caniato, F., Kalchschmidt, M., & Ronchi, S. (2011). Integrating quantitative and qualitative forecasting approaches: organizational learning in an action research case. Journal of the Operational Research Society, 62(3), 413-424.
Acknowledge