The weather has influenced not only on daily life but on all over the industries very closely and the previous researches had tried to prove the relationship. Especially, food service industry is influenced by the weather directly or indirectly and re...
The weather has influenced not only on daily life but on all over the industries very closely and the previous researches had tried to prove the relationship. Especially, food service industry is influenced by the weather directly or indirectly and reacted sensibly by the consumers so lot of researches has been going on. In our country, food delivery service is one of the special culture in the field of the food service industry.
Since later of 1990’s, the food delivery service has been growing up and the sum of market reached on about 12trillion Korean currency in 2015. But the related researches are not quite enough. Accordingly, in this study, we tried to specify the relationship between the weather effect and food delivery service. For analysis, we used the research data provided by SKT Big Data Hub and additionally daily weather data from the Weather Center.
There are temperature, humidity, rainfalls, wind velocity, cloud and daylight time, etc. for the examination of weather influence in the previous researches. But in this study, we try to add parameters that the consumers are feeling about the weather. Accordingly, we used the weather variation such as wind chill temperature, heat index and the temperature-humidity index, etc. Finally, we investigated that the explanation adding the above index is much better than not. Furthermore, we examined the detailed weather impact in each season and reached to the results. According to the previous researches, the season is defined with 3months duration such as Spring is from March to May, Summer from June to August, Autumn from September to November and Winter from December to February. Eventually, we examined the weather condition impacted on food delivery service are different and the variation to the food kinds. Also we find out not only statistical significance but economical significance for the fluctuation of food delivery order.
Finally, in this research, we try to predict the number of the food delivery order based on the weather data in 2015. So we utilized data mining technique such as artificial neural network, boosting and random forest for building prediction model. Also for the proof of predictive ability of the model, we tested the evaluation process once more with the data in 2016. 12. 26. At the result, we can see that the boosting predictive ability is most excellent among data mining technique.
Eventually, we proved the demand fluctuation of the food delivery order depend upon the weather condition by the seasons. Moreover there are lot of meaning that we are able to suggest reasonable marketing strategy to small sized food delivery industry by building up actual prediction model.