From IOUs to smart logistics, why is JD.com's big data more promising than AlphaGo?

A round of Go competition between man and machine has made the topic of artificial intelligence (AI) represented by AlphaGo a hot topic. At the event themed "'Understanding the Future' Entering JD - Changes Driven by Big Data " held at JD.com's headquarters on March 19, AlphaGo, which has brought great thinking to the future of mankind, has unsurprisingly become the most frequently mentioned term.

Essentially, the reason why AlphaGo can rewrite history is that it relies on having enough data to calculate and analyze when connected to the Internet. Looking at the current Internet industry, big data is one of the most concerned concepts in the industry.

If AlphaGo performs calculations based on chess records driven by big data, JD.com uses big data to promote the market and user experience, which is exactly the one closest to our real life.

How big is JD.com’s big data?

As we all know, as the industry's leading e-commerce platform, JD.com is constantly connected with users' consumption behaviors and has access to massive amounts of user data. But to what extent is JD.com's big data? Zhao Yihong, vice president of technology at JD.com Group, gave the following three sets of data.

First, JD.com's server scale has reached an industry-leading level of 10,000 units, which is very large in terms of number; second, these clusters run 200,000 report analyses every day and process 15 billion rows of data every day, which ultimately leads to JD.com's total data volume being very large.

According to Zhao Yihong, JD.com's total big data volume has now reached 100PB, and is growing by 1.5PB every day. This is only the current situation, and the possibility of a 100% increase again tomorrow cannot be ruled out.

Where are the applications of big data?

Zhao Yihong said that big data in JD.com is mainly divided into three strategic steps. In the past few years, a large amount of manpower and resources have been used to build JD.com's big data technology platform. On the basis of this technology platform, data insight, data value mining and discovery have been achieved, so that the value of big data has gradually been brought into play in JD.com's major businesses. Then, the next level will be the discovery and application of commercial value. How to combine big data with JD.com's business has become the key to future development.

In this regard, JD Group CTO Zhang Chen gave two important pieces of information.

The first point is to pursue high efficiency and low cost, such as JD.com's JIMI robot. Zhang Chen revealed that currently more than 50% of JD.com customer service conversations are conducted by JIMI with users. Pre-sales service JIMI's rating is often better than that of customer service. Customers need to understand the product before buying something. JIMI can give a relatively complete answer based on its understanding of user needs and data analysis. As a result, not only the user experience is improved, but also huge cost savings are achieved, especially before sales.

Another point is that JD Finance has become a big data credit system, making it possible to change users' traditional consumption habits. Based on the personal credit data model, JD Baitiao can decide whether to grant loans to users in a few minutes, which allows users to consume in advance while keeping JD within a certain safety factor. In addition, JD Finance can also lend to merchants. Through big data analysis, it can determine whether merchants can get loans and how much they can get. It can approve loans in three minutes, which helps merchants operate better on JD.

Accurate big data optimizes user experience

The huge volume of big data itself provides enterprises with enough reference benchmarks, but how to implement this data into actual services requires data analysis to make big data precise.

Taking the optimization of e-commerce delivery routes as an example, during the "Double Eleven" period, the distance walked by some warehouse pickers was almost the same as that of a marathon runner. Therefore, it was crucial to plan the route better. In more than 100 small and medium-sized warehouses, JD.com used big data analysis to make specific algorithms, which reduced the time for picking a single item from 22 seconds to 16 seconds.

In addition, many aspects of the distribution process, such as vehicle transportation, also require big data analysis. JD uses smart logistics to select the best distribution route, increase vehicle utilization, and reduce empty driving rates. Big data analysis, including weather conditions, optimizes the distribution route in order to deliver the goods to users as quickly as possible.

Before the goods are sold, accurate big data is also the support behind the personalized shopping experience. JD.com is now building a concept of a smart store. Everything is based on the understanding of users and accurate judgments based on user portraits. From a certain perspective, the basis for big data to truly generate value is to have very accurate user portraits, and to exclude data information that is conflicting. How to accurately identify users is the core of big data analysis.

Zhang Chen emphasized that compared with other companies, JD.com has the most complete, accurate and longest value chain data in China's e-commerce. While precise big data optimizes user experience, JD.com will also gradually open up big data and cloud services to create greater value with partners.

The number of variations in Go exceeds the number of atoms in the universe, and AlphaGo changed humanity's view of Go with just one move. JD's big data requires the use of 10,000 servers, and has promoted the e-commerce experience of tens of millions of JD users, from IOUs to smart logistics. In this sense, JD is closer to us and has a more imaginable future.

As a winner of Toutiao's Qingyun Plan and Baijiahao's Bai+ Plan, the 2019 Baidu Digital Author of the Year, the Baijiahao's Most Popular Author in the Technology Field, the 2019 Sogou Technology and Culture Author, and the 2021 Baijiahao Quarterly Influential Creator, he has won many awards, including the 2013 Sohu Best Industry Media Person, the 2015 China New Media Entrepreneurship Competition Beijing Third Place, the 2015 Guangmang Experience Award, the 2015 China New Media Entrepreneurship Competition Finals Third Place, and the 2018 Baidu Dynamic Annual Powerful Celebrity.