Foreword: This is a big data application article, which does not involve advanced technology. It is suitable for students who are new to big data to understand what big data can do. All data/news are from the public network.
Let’s talk about the latest news first. Didi, the fastest-growing company in recent years, is rumored to be laying off employees. It’s hard to tell whether the news is true or false. Online rumors: "Didi Chuxing may be undergoing the largest layoff since its establishment more than 4 years ago. Several sources of information told 36Kr that Didi is laying off employees recently; some people even informed 36Kr that various business lines of Didi In total, half of the employees will be "optimized" to leave. According to Didi's public statement, there are currently more than 6,000 people in various business divisions. Based on this calculation, this may mean that Didi may have to lay off 3,000 people? ". Does anyone who knows the details tell me the real situation?
This is not what I want to talk about today. Let’s talk about the big data application of Didi.
1. Didi is indeed a big data company
Data shows that the Didi platform generates more than 70TB of data (equivalent to 70,000 movies) every day, processes more than 9 billion route planning requests every day, and has an average daily positioning data of more than 13 billion. In 2015, the Didi Chuxing platform completed 1.43 billion orders, which is equivalent to the average person in China using Didi to drive once; the cumulative mileage reached 12.8 billion kilometers, equivalent to 290,000 laps around China, and the cumulative driving time was 490 million hours, equivalent to 56,000 years of driving day and night. This year, the daily peak orders of Didi platform exceeded 20 million, and the annual order volume in 2016 will far exceed last year.
2. Didi's intelligent transportation cloud
There are two sources of data for Didi, one is data from its own software and mobile phones, and the other is to cooperate with the government to obtain public data.
On the Didi intelligent transportation cloud platform, through the collected travel big data, regional heat map, OD data analysis, urban capacity analysis, urban traffic travel forecast, urban travel report, and dynamic timing of signal lights can be realized. Public travel services, such as real-time traffic conditions, real-time public transport, ETA, and urban capacity supplementation, play a huge role.
In the future, by integrating sensor data, static road data, road events and other data with Didi's OD data, driver data, GPS trajectory data, and capacity data, it will provide better services for the entire city's transportation.
3. Didi's big data application
Let's take a look at what Didi has done with big data.
1) Through accurate analysis and prediction of big data, the estimated cost and actual cost are consistent.
2）使用热力图提前预测需求，蜂窝动态调价，提升整体成交率。
3）智能拼车，通过虚拟站点设计，撮合不同地点乘客拼车。
4）此外，滴滴通过和政府合作，和城市一起实现智能交通。（目前已知滴滴和沈阳、武汉都签署了战略合作协议。目前看在这个能力目前应该是还没有实现，规划中。）
例如智能信号灯控制，“通过数据模型算出整个区域的车流量情况，理想情况下，比如让车在主干道上通行效率更高，可以靠区域的红绿灯协调实现。”
以上简单聊下滴滴的大数据应用，知道更多详情的同学留言一起讨论吧。
我的新作《大数据架构详解：从数据获取到深度学习》一书，已由电子工业出版社出版，当当热卖榜排名第5，京东热卖榜第3。另外一个好消息是：由于销售太火爆，刚销售一个月，编辑就启动了重印。
此书京东，淘宝，当当，亚马逊全网有售，有兴趣的同学直接上京东，淘宝，当当，亚马逊 搜索书名，详细了解：
微信扫一扫
关注该公众号