Home Edge Computing Big data utilization examples: Big data utilization that has already begun

Big data utilization examples: Big data utilization that has already begun

Big data analysis that feels like something that is not yet familiar in the advanced technology field, but it is gradually beginning to be utilized. Here, we will introduce examples of utilization of big data analysis and the merits of it, and introduce examples of IoT and industrial fields in particular.

The Eye of Big Data Utilization

The purpose of analyzing big data is to find some law and correlation from seemingly disorderly and non-lawless data. From there, we are trying to increase sales and improve operational efficiency by deriving new knowledge.

Decision-making based on human experience and intuition is highly uncertain and hits when it hits, but it often comes off. In addition, it takes a long period of time to acquire the intuition to make decisions. In addition, as customer needs are diversifying and the pace of market change is getting faster and faster, it is very likely that the experience and intuition cultivated over many years will not be passed in a short period of time. In this situation, it is necessary to quickly provide new knowledge by analyzing large amounts of data without human hands.

In addition, big data analysis is considered useful for grasping laws and some trends based on human preferences and sensibilities that were previously difficult to quantify. This means that new knowledge can be discovered from data that was previously excluded from the difficulty of responding.

Examples of how to use familiar big data analysis

Let’s take a look at real-world big data use cases.

Product display at vending machines

Product displays are often used as a good example of creating laws from big data analysis. At a beverage manufacturer, we changed the product display of a “Z-shaped” vending machine that arranged the best-selling products from above, which was commonplace until then, so that the products we were focusing on were placed below. This is the insight derived from installing an eye tracking camera (eye tracking camera) that reads the line of sight in a vending machine and analyzing big data called “shopper’s line of sight”. As a result, the company achieved a sales increase of several percent.

Many vending machines display product samples higher than buyers. This is because it is often fixed on a concrete pedestal to prevent falls during earthquakes. Certainly, it seems natural to say that it is easier to see in that position at a slightly lower position than the height of your eyes. However, ideas that overturn the custom that is already considered common sense, such as the display of the “Z-shaped”, are difficult for humans alone to come up with.

Understand demand fluctuations and use them for purchasing and product development

One of the representative applications of big data analysis is demand forecasting. In one conveyor belt sushi chain, ic tags are attached to the plates to figure out which dishes sold and how much. When the clerk hold the reader over the plate at the time of accounting, data can be acquired instantly. This data is several hundred million annually, and by analyzing such big data, we grasp sales and seasonal demand fluctuations, and use it for adjustment of purchase volume and product development.

Analyze customer trends for management decisions

In some cases, it is used for sales analysis and inventory management at convenience stores. One convenience store found that 60% of the sales of a particular product were supported by only 10% of heavy users. This is the knowledge obtained by analyzing big data information on electronic money settlement. This convenience store continues to sell certain products because they determine that they have an avid fan.

An important source of revenue for convenience stores is “next to buy” purchased together with the desired product. Even if a product that at first glance seems to have to end sales due to sales growth, you can find new value as an important product that leads to attracting customers by knowing that you have enthusiastic fans.

Human senses also deciphered as big data

On an underwear sales site, we succeeded in quantifying slightly different size variations in underwear manufacturers by analyzing the returned underwear and customer information with big data. This led to a system in which when a buyer enters his or her size, only products from the manufacturer that fit that size are displayed.

It can be said that this was able to analyze big data on the fit and comfort of underwear in different different people and present the best products to each customer. Since the fit and comfort of underwear belong to human senses and sensibilities, it has been considered difficult to quantify in general, but this information can also be analyzed as big data depending on the ingenuity.

Examples of utilization of big data analysis in industrial and IoT fields

Let’s take a look at examples of use in the industrial and IoT fields.

Grasp and forecast road conditions in real time

There is a case where the traffic information is analyzed by giving the communication function to the car. By collecting driving information using a car with communication functions, a so-called connected car, and analyzing it with big data, it is possible to obtain traffic jam information and accident information. This data can be used to avoid traffic jams and prevent accidents.

As a mechanism, it is determined that there is no traffic jam if the rotation speed increases or is constant by sensing how many revolutions the tire rotated per second. In other words, the road situation is judged from the change in vehicle speed. Furthermore, by installing a GPS receiver in the car, sensing the position information, and performing big data analysis in conjunction with the change in individual vehicle speed, it is possible to determine how much traffic jam is occurring or likely to occur in which region. In this case, since the information from the automobile changes from moment to moment in real time, real-time processing is required for big data analysis.

Efficient ship operation

In the same way, there have been cases where the operation of ships has been streamlined by analyzing big data obtained from sensing. A merchant shipping company attaches sensors to ships to collect engine speed, running speed, and ocean current speed. We have succeeded in analyzing this as big data and creating an optimal flight plan. When we actually operated using this plan, we achieved a 10% reduction in fuel consumption. From the general public’s view, it seems that it is not a big deal at 10%, but merchant shipping companies operate a large number of ships, and there are many cases where the voyage at one time is long. 10% of these results can be said to be the effect of major cost reductions.

Tribute to optimization of mining

In some cases, it contributes to the efficient operation of construction machinery such as shovel cars and bulldozers. Construction machinery of a construction equipment manufacturer has sensors compatible with IoT. By using this sensor to monitor location information and operation status and perform big data analysis in real time, it can be used to improve operation efficiency and predict failures. This is mainly useful when mining in overseas mines. Overseas mines often have a very wide mining range, and it is necessary to choose the shortest distance possible and carry empty loads less frequently. Therefore, optimization of mining is realized by sensing the start-up time of the engine, the travel distance, the direction of travel, etc. and analyzing this big data.

Big data analysis will be more real-time oriented

Above, we have seen examples of how big data analysis can be used. As you can see from these examples, the need to perform big data analysis in real time will increase more and more. Edge computing is likely to be highly compatible in terms of speed, especially for applications that require speed in the production field. If you are thinking about big data analysis in this production field, the introduction of edge computing is also an option.

Related Posts