Business expectations for demand forecasting and dynamic pricing
I think that the utilization of big data will progress in various fields going forward. It will further improve the convenience of our lives.
On the other hand, it is necessary to know the mechanism used, think about how to use it, and discuss it.
On EC sites, etc., a recommendation system that collects customer data and displays individual recommendations to customers based on data has already been put into practical use. However, there are also efforts to utilize big data and AI in a larger framework, for example, demand forecasting and dynamic pricing.
It seems that demand forecasting is already being introduced in the apparel industry.
For example, a major company adopts various factors such as consumer trends, economic conditions, and climate to make a rough demand forecast, and then produces fabric based on that forecast. If you secure the fabric, you can quickly respond to the demand for individual products.
In other words, it is still difficult to forecast demand for individual products using AI, but systems that are easy to handle it have been developed in the apparel industry.
It is also expected that more sophisticated demand forecasting will enable more effective dynamic pricing.
For example, even today, airfares and accommodation charges in the travel industry are expensive during the busy season and low during the off-season. For that purpose, demand forecasting based on customer data and accumulated past data is performed by AI.
However, it can be difficult to introduce such a system in a highly competitive industry, as overpriced settings can lead the industry to lose customers. Compared to that, it is relatively easy to introduce EC sites if there are a few limited major companies.
However, customers who want to buy when the price is cheaper will have to constantly check the price fluctuations. Customers may also be dissatisfied if they find that the price of the item has dropped further after they purchased.
Such stress can lower customer satisfaction and lead to losing customers. This is difficult for businesses to judge.
The ability to relativize information and capture it objectively is important
Utilization of data analysis by AI, etc. is also expected in the accounting field. For example, detection for fraudulent accounting based on accounting information.
It is often said that the use of AI deprives people of their jobs. However, checking accounting information is a difficult task for accountants, so the use of AI is highly expected in terms of supporting this task and reducing the burden.
Therefore, accountants and others are also participating in the development of algorithms that detect fraud patterns from accounting information.
In this way, the utilization of data by AI is being promoted in various fields. However, businesses that are thinking about introducing AI systems should not think of AI as a magical system that can provide an answer for everything.
For example, if you try to use AI effectively, the more data you have, the more accurate the calculation will be. Therefore, it is necessary to collect as much data as possible in addition to the customer data and product data that your company has.
And then, not only costs for hardware and software, but also costs for data collection and its management have to be considered. It is uncertain how much profit can be expected to increase against such total costs. In other words, at the current state of technology, AI is by no means a cost-effective system.
I think it is appropriate to flexibly incorporate the necessary parts of AI in order to support a human task.
On the other hand, from the perspective of consumers, as is often said, it is important to increase literacy.
For example, the idea that it’s okay if you can exchange some personal information for convenience or benefit is just one value and way of thinking. Nevertheless, we need to consider the risks of offering personal information and pay attention to how your personal information is managed and utilized.
Also, if businesses proceed with demand forecasting and dynamic pricing, they will not know what the list price is. Then consumers may perceive that it’s a benefit based only on the information they obtained.
In short, consumers can also be manipulated in marketing. In that sense, literacy is also important.
Also, if a system that detects fraudulent patterns from accounting information is created, some people will devise algorithms that are not detected as fraudulent patterns. In other words, it is also necessary to judge whether the obtained information is forged information or not.
People are more likely to be tempted by the benefits and information they want to believe. I think that the ability to relativize information and capture it objectively, that is, literacy, will become a more important skill for us as more big data are utilized by AI.
* The information contained herein is current as of December 2020.
* The contents of articles on Meiji.net are based on the personal ideas and opinions of the author and do not indicate the official opinion of Meiji University.
* I work to achieve SDGs related to the educational and research themes that I am currently engaged in.
Information noted in the articles and videos, such as positions and affiliations, are current at the time of production.