Author: Jacques ter Huurne


Forecasting is the process of making predictive statements based off previously analysed data of events that haven't occurred yet.


There are various different kinds of forecasting ranging from weather and pattern forecasting in meteorology and science to sales and demand forecasting in business. Forecasts are divided into the time intervals of which they are predicting over, long term projections or short-term.

Research methods are also confined to two different types in forecasting: quantitative and qualitative. Otherwise known judgemental methods, qualitative forecasting rely on expert opinion to make predictions on future event based on experience. These methods are suitable for mid to long ranged forecasting tasks, examples of these include market survey (of the target audience), historical extrapolation, the Delphi method amongst others.

Explanation and application

Forecasting is similar to extrapolating data. Except that forecasting is using data to estimate what events will be happening in the future, not numbers, like extrapolating data. The biggest problem for supermarkets or other companies with forecasting, is the fact that there is a big uncertainty that comes with forecasting, since we are not able to tell the future.

Current examples in Supermarkets include efforts by certain chains like Tesco to predict and influence buying patterns based on data analytic patterns in weather forecasting. Supermarket chains like Tesco and Otto (in Germany) build these computer models that can build future demand for products from store to store based on the local area's weather forecasting. This way they can save costs on stocking the right amounts of right the kinds of products during each season in each region. The forecasting program relies on a 100 terabyte data warehouse using sophisticated data analytics software to handle massive amounts of data collected from weather stations and satellites over the past 15 years and compiling it into a predictive model that can work with contemporary data as well. But what does this mean for consumers? Well, when the temperature rises from 20 C˚ to 24˚C for instance, the sales for hamburgers may rise by 42 percent, according to Tesco researchers. A 10˚ rise in temperature may mean that more people will want to buy coleslaws and barbecue sauce, but demand for green vegetables and cauliflower would lower by 25 percent. It's these clues based on slight temperature and weather changes that can help drive sales of certain products for certain occasions and even get supermarkets to campaign certain products during weather change. For example Tesco research forecaster Ian Michaelwaite says that, ‘Our behaviour is interesting. For example people think we buy a lot of ice cream in summer, but actually ice cream sales plateau around 25C in the south and then its frozen lollies that go up a lot. In Glasgow, it’s the same at about 19C because they are used to it being much colder.’

He said: ‘Weather has the biggest impact on sales after the general state of the economy. Hot weather is not always good, shops want that in summer so they can stock up on barbeque food and summer dress, and then in about six weeks from now they would like the weather to turn noticeably colder.'

Links to Social and Ethical Issues

Please note that not all issues need to be addressed. Please add the URL or source of any examples to support your suggestion. It may be helpful to RANK the issues in the THIRD column.

Social & Ethical Issue
Examples that specifically link to the concept and/or definition in the Case Study
1.1 Reliability and integrity
This is definitely an issue of reliability and integrity of the weather data gathered and
processed by the program and predictive models set up by forecasters. Accuracy
of the data can greatly influence how effective of a forecast it will be and how likely
the desired outcome is met. Supermarkets will want to be able to keep track of
weather forecasts closely else they come across reported weather change that
doesn't really turn out to be the case at all.

1.2 Security

1.3 Privacy and anonymity

1.4 Intellectual property

1.5 Authenticity

1.6 The digital divide and equality of access

1.7 Surveillance

1.8 Globalization and cultural diversity

1.9 Policies

1.10 Standards and protocols

1.11 People and machines

1.12 Digital citizenship

References and resources

"Forecasting Principles." Forecasting Principles. Marketing Department, July 2006. Web. 17 Oct. 2014.

Erhun, Feral. "Enterprise-Wide Optimization of Total Landed Cost at a Grocery Retailer." Pubsonline. Pubsonline, n.d. Web. 17 Oct. 2014.