Logistics big data data is through the massive logistical data, namely transport, storage, handling, packaging and distribution processing and other logistics sectors involved, information, and tap new added value, can improve transportation and distribution efficiency through big data analysis reduce logistics costs and more effectively meet customer service requirements.
1. The role of the large data stream
Logistics big data applications play an important role for the following three aspects of the logistics enterprises,.
1) improve the intelligence level of logistics
By tracking and analyzing data on the logistics, the logistics of large data applications can make decisions and recommendations of intelligent logistics enterprises according to the situation. In the Logistics Decision, big data technology applications involving competitive environment analysis, logistics supply and demand matching, optimization and configuration of logistics resources.
In the analysis of competitive environment in order to achieve maximum benefits, it is necessary to conduct a comprehensive analysis of competitors, predict their behavior and movements, to understand, or at a particular time, it should be the partner of choice in an area.
Logistics to match supply and demand, the need to supply and demand analysis stream specific period, a specific area, so that a reasonable distribution management. In the optimization and allocation of logistics resources, mainly related to transport resources, storage resources. Logistics market has a strong dynamic and random, real-time analysis needs changing market conditions, the current logistics needs to extract information from vast amounts of data, at the same time and will be configured to optimize the allocation of resources in order to achieve a reasonable logistics resources use.
2) reducing logistics costs
Due to transportation, storage facilities, cargo packing, processing and handling other aspects of information exchange and sharing requirements are relatively high, so you can use big data technology to optimize the delivery route, a reasonable choice logistics center address, optimize warehouse storage spaces, thus greatly reducing logistics costs and improve logistics efficiency.
3) improve the level of customer service
With the rapid expansion of online shopping population, more and more attention to customer experience of logistics services. Through data mining and analysis, and rational use of the results of these analyzes, logistics companies can provide customers with the best service, to provide all the information logistics operations during the delivery of goods, and further consolidate the relationship between the customer, increase customer trust, develop customer stickiness, reduce customer turnover.
2. Logistics Big Data Applications
According to the characteristics of the logistics industry, large data applications mainly in the car matched the cargo transportation route optimization, inventory forecasting, equipment repair forecasting, collaborative supply chain management.
1) goods vehicles match
By analyzing large data capacity pool, you can produce a good match between standardization and individual needs of the professional capacity of the capacity of the public, at the same time, combined with enterprise information systems will be fully integrated and optimized. By accurate portrait of the owner, the driver and tasks, enabling intelligent pricing, for drivers intelligent recommendation to the tasks assigned task and delivery drivers and so on.
From the client side, big data applications will be based on mission requirements, such as cars, delivery a few kilometers, when the distribution is expected to grow, additional services, such as automatic calculation of price and capacity to match the most qualified driver, the driver received a task in accordance with the requirements of customers high-quality service. In terms of drivers, big data applications can be automatically matched according to the driver's personal circumstances, quality of service, the right of free time for his mission, and intelligent pricing. Based on large goods vehicles to achieve efficient data matching, not only can reduce the loss of Kongshi bring, but also reduce pollution.
2) optimization of transport routes
Through the use of big data, logistics and transport efficiency will be greatly improved, large data between logistics enterprises will bridge the communication, logistics and vehicle traffic will also be the shortest path, optimization customized.
UPS US companies use big data to optimize delivery routes, delivery personnel do not need to think for themselves whether the optimal distribution path. UPS systems with large real-time data analysis of 20 million possible routes, three seconds to find the optimal path.
UPS through big data analysis, the provisions of trucks can not turn left, therefore, the UPS driver will prefer to circle around, do not turn left. Based on past data show that since the implementation of policies to avoid left turns, UPS truck in driving distance at 204 million reduction in the premise, sent out more than 350,000 parcels.
3) inventory forecast
Internet technology change and business model has brought change directly from the producer to the customer's supply channels. Such a change, time and space are two dimensions to create new value of the logistics industry has laid a good foundation. Large data structures and techniques may optimize inventory storage reduce inventory costs.
The use of big data analytics product category, the system will automatically be used to break down used to promote drainage and commodities; at the same time, the system will automatically modeling and analysis based on past sales data in order to determine the current safety stock of goods, and in a timely manner to Chu warning, rather than to predict the current inventory based on sales in previous years. In short, the use of Big Data technologies can reduce inventory stock, thereby improving the utilization of funds.
4) prediction apparatus Repair
American UPS company from 2000 began to use predictive analysis to detect the size of their own nation's fleet of 60,000 vehicles, so that we can promptly be defensive repairs. If the car breaks down on the road, the loss will be very large, because that would need to send another car will cause delays and burden of reloading, and consume a large amount of manpower and material resources.
Previously, UPS every two or three years will be on the parts of the vehicle were replaced regularly, but this method is not very effective, because some parts and there is nothing wrong with it was replaced. By monitoring all parts of the vehicle, UPS now only need to replace the necessary parts, saving millions of dollars.
5) Collaborative Supply Chain Management
As supply chains become more complex, the use of Big Data technologies can be quickly and efficiently to maximize the value of enterprise data integration of all business planning and decision-making, including demand forecasting, inventory planning, resource allocation, equipment management, channel optimization, production planning, material requirements and procurement plans, which will revolutionize the enterprise market boundary, portfolio, business model and operation mode.
Good supplier relationships are among the eradication of key suppliers and manufacturers do not trust the cost. Inventory and demand information interaction of the two sides, will reduce production because of shortages caused by the loss. By storing resource data, transaction data, supplier data, quality data up to track and analyze supply chain during the execution of efficiency, cost, can control product quality; through mathematical models, optimization and simulation technology overall balance orders, the relationship between production, scheduling, inventory and costs, find the optimum solution to ensure an orderly and uniform production process, ultimately splitting decomposition best supplies and production orders.
3. Amazon Logistics big data applications
Amazon is the largest global online retailer of varieties of goods, adhere to self logistics direction, which will be integrated logistics and large data closely linked, in order to achieve greater value in marketing. Because Amazon has improved, optimized logistics system as a guarantee, it can be logistics as a means of promotion, and the ability to strictly control the logistics costs and effectively organize the logistics process operation.
Amazon first in the industry to use big data, cloud technology and artificial intelligence to manage storage and logistics, the introduction of innovative predictive allocation, inter-regional distribution, cross-border distribution and other services.
1) orders for big data applications and customer service
Amazon a complete end to end service five major categories: browsing, shopping, warehouse distribution, delivery and customer service.
① View
Amazon based big data analysis techniques to accurately analyze customer needs. Browsing through the customer's system records the history, the background will follow the interest of customer inventory on their nearest Operations Center, which facilitate customer orders.
② Shopping
No matter which corner of customers, Amazon can help customers to quickly order, you can quickly know their favorite products.
③ cartridge with
Amazon operations center fastest to complete the entire process orders within 30 minutes. Large data-driven warehousing operations very efficient orders, order processing, picking all the process fast, fast packing, sorting and other data driven by the large and full visualization.
④ delivery
Amazon's logistics system will be based on the customer's specific needs time science stowage, distribution adjustment program for precise delivery within the time frame a user-defined. Amazon also can predict big data, advanced shipping, win an absolute competitive edge.
⑤ Customer Service
Amazon using the large data-driven customer service, create a technical system to identify and anticipate customer needs. According to the user's browsing history, order information, calls the problem, customized to push to the different user self-service tools, big data can ensure that customers can call anytime to the corresponding customer service team.
2) intelligent storage management technology
Amazon operations center in the world, this time from warehouse has been using big data technologies.
① storage
Amazon uses a unique purchase storage monitoring strategies, based on their past experience and gather all the historical data to understand what kind of category easily broken, where the bad, and then to its pre-packaged. This is the value-added services in receipt link provided.
② commodity measurement
Cubi Scan Amazon's new instruments have a small storage volume and the volume of goods to measure the length and breadth of these storage and optimize product information. This gives the supplier provides a great convenience, customers do not need to measure their own new products, this can greatly enhance the speed of new products on-line. The database stores data in the Amazon, nationwide shared, so other warehouse can directly use these background data for subsequent optimization, design and regional planning.
3) picking smart and intelligent algorithms
Amazon use big data analytics to achieve a smart pick, mainly used in the following areas.
① intelligent algorithm driven logistics operations to ensure optimal path
Data algorithm will give Amazon's big data platform for logistics optimization of each person randomly picking his path. The system will tell which employees should go to cargo picking and ensure minimal path after the completion of the entire election. By calculating such a smart and intelligent recommendation can pick traditional mode of travel path by at least 60%.
Complex operating methods ② book warehouse
Book warehouse uses an enhanced version of surveillance, we will try not to limit the product are similar to those in the same cargo space. Purchase volume book of great batch of "Amazon through analysis of the data found interspersed placed guarantee higher per employee picking out the task.
③ operating strategy of selling products
According to Amazon's big data background, can demand to know which items are relatively high, and then put them in the delivery area from the more recent places, and some are on the shelves, some care is on the beat position, this reduces the weight-bearing walking distance employees.
4) Intelligent Shuffle storage
Amazon random storage technology is an important operation, but not just any random store to store, but there is a certain principle. Random access memory to consider selling merchandise and non-merchandise sold, but also consider the FIFO principle, while random access memory also has an important relationship with the best path.
Random shelves is a major feature of Amazon's operations center, to achieve the best storage must attack. Seemingly messy, actually chaos in the orderly. Chaos is one that can break down barriers between the category and category, we can put them together. Ordered label refers to the location is that it's GPS, the cargo space inside all of the goods actually inside the system is in its place, to be very precise in what it recorded in the area.
5) a smart and intelligent allocation binning
Amazon intelligent warehouse and intelligent allocation of points has a unique technical advantages, the allocation of positions in more than 10 parallel Amazon China is entirely carried out at precise drive supply chain program, which implements the intelligent sub-warehouse, stocking and predictive nearby allocation.
The country in various provinces and cities including transport route between the major operators in the deployment center to ensure that inventory has been allocated in advance to the closest to the customer's operations center. Intelligent allocation of the entire national transportation network is well supported the concept of parallel positions, as long as the goods across the country the following single user can buy, this is the big data system to support the national transportation network to allocate adequate performance.
6) accurate inventory forecasting
Amazon's intelligent storage management technology enables continuous dynamic inventory, inventory forecasting accuracy rate of 99.99%. In peak periods, Amazon can be done through big data analytics accurate forecasting of inventory requirements, and be ready in picking planning, capacity allocation, distribution and end, so as to balance the operational capacity of orders, greatly reducing the risk of warehouse explosion .
7) Visualization ,, package tracking work orders
Amazon to achieve a global visualization of supply chain management, inventory can see from across the Atlantic in China. Amazon platform allows domestic customers, partners and staff monitor the whole Amazon goods, wrapping position and order status. From the reservation to the front of the store receipt to internal management, inventory allocation, picking, packing, distribution and then shipped to the customer, the entire process chain, each process has its supporting data, and through system visualization of its management.
4. International Logistics big data applications
DHL use big data to speed up their response rate by analyzing customer data do accurate service; UPS adjust the distribution strategy saves a lot of fuel costs through large data; Fleet Risk Advisors can do full monitoring of fleet management, even aware of the driver's psychological changes.
1）DHL
DHL Express shipping company, Express truck was specially adapted to become Smart Truck, and equipped with Motorola's XR48ORFIO reader. Whenever a transport vehicle loading and unloading goods, the RFID sensors on-board computer will goods information uploaded to the data center server, the server dynamically calculates the latest sequence and the optimal distribution path after the update data.
In addition, in transit, telematics databases updated in real time based on real-time traffic conditions and GPS data distribution path, so that more accurate pickup and delivery of orders received at any time to make a more flexible response, and provide customers accurate information about the pickup time. As shown in Figure 1.
FIG 1 DHL large data stream applied
DHL By operations on large data collection terminal, the full realization of visual monitoring and scheduling the optimal path, while closely to each operator node. In addition, the customer has a Crowd-Based mobile applications can be updated in real time their location or are about to reach the destination, DHL parcel delivery personnel can receive real-time location information of the customer, prevent delivery failures, or even updates on demand delivery destination.
2）FedEx
FedEx FedEx package allows active transmission of information. To achieve near real-time feedback through a flexible sensor (such as SenseAware), including temperature, light, and the location, so that customers can know at any time which the wrapping location and environment, and the driver can also be modified in order directly car logistics information.
In addition, FedEx is working to promote a more intelligent delivery of services; real-time updates and understanding in the case allowed the customer's geographical location, the parcel more quickly and accurately reach the hands of customers. FedEx may in the future based on historical data collected and real-time incremental data, FedEx solutions to solve more problems by big data, so as to enhance competitiveness. as shown in picture 2.
FIG 2 FedEx large data stream applied
3）FleetBoard
FleetBoard committed to large data processing provides remote fleet management information technology solutions for the logistics industry users, data acquisition and process monitoring, including driving the driver's driving operation, vehicle temperature, door open and other details. Terminal to establish contact on the vehicle, the data exchange via a mobile communication system FleetBoard server.
Logistics company or fleet managers can directly access the GPS and a number of other real-time data, such as the direction of travel, parking / travel time and loading / unloading and other information. In addition, by calculating the driver's hard acceleration, braking times, the economy and the idle speed zone travel time and length information, you can directly help the driver find the problem and improve driving commands to improve. FleetBoard large data stream applied as shown in FIG.
FIG 3 FleetBoard large data stream applied
For users of cold chain transport, FleetBoard has real-time temperature monitoring of refrigerated trucks of specialized data management system, the door is open, etc., automatically sends a warning message to the phone or e-mail.
4）Con—Way Freight
Con-Way Freight provides LTL transportation, third-party logistics and bulk cargo transportation services, covering 18 countries on five continents of North America and the United States.
Con-Way Freight solved by using big data solutions allows the system to integrate real-time incremental data, and quickly obtain accurate answers through inquiry and processing of unstructured data.
Ad-Hoc 系统使得公司可以定义需要监控的配送流程，预测商业活动内部和外部因素的影响，以及为 CRM 和营销计划提供消费者划分，甚至可以定位到任何一位客户，实时分析送达率和具体的货运损失等信息。而 Score Carding 系统能够将原定目标和实时表现进行对比，使 Con-Way Freight 能够随时根据对比结果全面调整和提高运营表现。如图 4 所示。
图 4 Con-Way Freight 物流大数据应用
Con-way Freight 高管能够通过大数据解决方案快速得出准确的数据报告，做岀恰当及时的运营决策。
5）C.H.Robinson
C.H.Robmson 第三方物流公司拥有全美最大的卡车运输网络，却没有一辆货车。它用 1.5 亿美元的固定资产，创造了 114 亿美元的收入、4.5 亿美元的利润。它的新生始于 1997 年的商 业模式变革，主动放弃了自有货车，建立了专门整合其他运输商的物流系统，通过系统对社会 资源进行整合建立新的平台经济。如图 5 所示，C.H.Robmson 的平台模式由 3 部分构成：TMS 平台，用来链接运输商；“导航球” Navisphere 平台，用来连接客户；做支付的中间账户， 同时提供咨询服务。2012 年，支付服务带来大约 5 亿美元的净收入，咨询服务带来了 12 亿美 元的收入。
C.H.Robinson 通过系统的两大平台：导航球（Navisphere）和 TMS 平台，来对接客户群和运输商，沉淀形成的大数据库可支持 C.H.Robinson 的增值服务。
图 5 C.H.Robinson 物流大数据应用
6）FRA
FRA（Fleet Risk Advisors）为运输行业提供了预测分析和风险预防或补救解决方案。FRA 根据历史数据和实时增量数据可得出司机工作表现模型和若干预测模型，能够准确地预测可避免的事故、员工流动等问题。
For example, according to the driver real-time performance fluctuations, predict driver fatigue and rostering arrangements, to provide customers with a reasonable solution to improve driver safety factor, also can predict the possible risks based on real-time status of the driver and the vehicle and provide timely preventive or remedial solutions. FRA data forecast model using a large achieved good results, as shown in FIG.
FIG 6 FRA large data stream applied
FRA draw some predictive model driver performance through big data solutions that solve the problem of the personnel department of the accident rate and the movement of persons and so on.
Recommended Learning Catalog: Application of the logistics industry in large data