登录查看更多内容

Last updated on 2024年11月9日

What are some best practices for using expert opinions for demand forecasting?

由人工智能和领英社区提供技术支持

Demand forecasting is the process of estimating the future demand for a product or service based on historical data, market trends, and other factors. However, sometimes data is not available, reliable, or sufficient to support quantitative forecasting methods. In such cases, you may need to rely on expert opinions or judgmental forecasting to make informed decisions. Expert opinions are the subjective estimates or judgments of experienced and knowledgeable people who have insights into the demand situation. But how can you use expert opinions effectively and avoid bias and error? Here are some best practices for using expert opinions for demand forecasting.

此文章中的业界达人

由社区从 6 条内容中精选。了解更多

Dave Stuart

Business Owner at Stuart House Consulting

1 Choose the right experts

The quality of your expert opinions depends largely on the selection of the experts. You want to choose people who have relevant domain knowledge, experience, and credibility, but also diverse perspectives and backgrounds. Avoid selecting experts who are too close to the product or service, have personal agendas, or are influenced by peer pressure or groupthink. You may also want to consider the number of experts, depending on the complexity and uncertainty of the demand situation. Generally, more experts can provide more information and reduce individual bias, but they may also increase the cost and time of the forecasting process.

添加您的观点

Dave Stuart

Business Owner at Stuart House Consulting
举报内容
The methodology of choosing experts shown sounds very wide and indiscernible. I’d do a fast (EDA) forecast of each relevant expert to get a rough outcome of expectations (growth rates, tenor shape, etc). That will help guide which way to proceed with un-achievable data.

已翻译

赞
Gazi Sanaul Hasan

SPL DECATHLON | EX P&G | EX SIA | EX Shopee | Micro Masters Candidate MITx | B.Sc. in Logistics & Supply Chain | Six Sigma Yellow Belt
举报内容
For demand forecasting with expert opinions, first select knowledgeable experts across key areas like sales and market trends. Use structured methods, such as the Delphi method, to gather and refine their input systematically. Quantify opinions by combining them with historical data for objective analysis. Regularly update forecasts to account for market shifts, and measure forecast accuracy to assess expert reliability. Encourage collaboration among experts to reduce biases, ensuring their insights align closely with actual demand trends.

已翻译

赞

2 Define the forecasting problem

Before you ask the experts for their opinions, you need to clearly define the forecasting problem and communicate it to them. You should specify the product or service, the time horizon, the level of aggregation, the units of measurement, and any assumptions or constraints that apply. You should also explain the purpose and scope of the forecasting exercise, and how the expert opinions will be used and evaluated. By defining the forecasting problem, you can ensure that the experts understand what you are asking them and provide consistent and relevant opinions.

添加您的观点

Dave Stuart

Business Owner at Stuart House Consulting
举报内容
I’d even go back further and say spend more time defining the forecast/business problem (unknown). Defining the problem correctly often leads directly to the data analytics solutions required.

已翻译

赞

3 Collect and aggregate the expert opinions

There are different methods to collect and aggregate the expert opinions, depending on the level of interaction and feedback among the experts. For example, you can use surveys, interviews, Delphi method, nominal group technique, or scenario analysis. Each method has its advantages and disadvantages, and you should choose the one that suits your forecasting problem and resources. When you collect and aggregate the expert opinions, you should aim to elicit honest, independent, and realistic opinions, and avoid leading or influencing the experts. You should also document the sources, methods, and results of the expert opinions for transparency and accountability.

添加您的观点

Dave Stuart

Business Owner at Stuart House Consulting
举报内容
Do not use expert opinions to decide important decisions. Ever. Experts know the right answers, only if you ask the right questions. For every expert, there is an equal and opposite expert.

已翻译

赞

4 Analyze and adjust the expert opinions

After you collect and aggregate the expert opinions, you should analyze them for validity, reliability, and accuracy. You should compare the expert opinions with historical data, market trends, or other sources of information, and look for any patterns, outliers, or discrepancies. You should also check for any bias or error that may affect the quality of the expert opinions, such as overconfidence, anchoring, availability, or confirmation bias. Based on your analysis, you may need to adjust the expert opinions to correct for any bias or error, or to align them with your forecasting objectives and assumptions.

添加您的观点

Dave Stuart

Business Owner at Stuart House Consulting
举报内容
This cannot be done in an unbiased way. Experts have great volatility of relevance and predictability over time. They help guide your direction of analysis but definitely not the actual analysis.

已翻译

赞

5 Combine the expert opinions with other forecasting methods

Expert opinions are not a substitute for quantitative forecasting methods, but a complement to them. You can use expert opinions to supplement or modify your quantitative forecasts, or to generate scenarios or ranges of possible outcomes. For example, you can use expert opinions to adjust your quantitative forecasts for factors that are not captured by historical data or models, such as new products, competitors, regulations, or events. You can also use expert opinions to create scenarios or ranges of demand based on different assumptions or conditions, such as best-case, worst-case, or most-likely scenarios. By combining the expert opinions with other forecasting methods, you can improve the accuracy and robustness of your demand forecasts.

添加您的观点

Dave Stuart

Business Owner at Stuart House Consulting
举报内容
I would never do this. I cannot define what/who is the supposed expert. And if we could, why would I be doing the analysis since the expert is available. “Expert analysis” is inherently tautological.

已翻译

赞

6 Monitor and update the expert opinions

Expert opinions are not static, but dynamic and evolving. You should monitor and update the expert opinions periodically, or whenever there are significant changes in the demand situation or environment. You should track the performance of the expert opinions against actual demand, and measure their accuracy and error. You should also solicit feedback from the experts, and ask them to revise or update their opinions based on new information or evidence. By monitoring and updating the expert opinions, you can keep them relevant and reliable, and enhance your demand forecasting process.

添加您的观点

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?

添加您的观点

Demand Forecasting

+ 关注

给文章评分

我们借助人工智能创建了此文章。您认为这篇文章怎么样？

很棒不太好

举报此文章

查看全部

What are some best practices for using expert opinions for demand forecasting?

1

2

3

4

5

6

7

1 Choose the right experts

2 Define the forecasting problem

3 Collect and aggregate the expert opinions

4 Analyze and adjust the expert opinions

5 Combine the expert opinions with other forecasting methods

6 Monitor and update the expert opinions

7 Here’s what else to consider

Demand Forecasting

给文章评分

感谢您的反馈

更多Demand Forecasting相关文章

更多相关阅读内容