In recent years, all walks of life have been talking about digital transformation, and the public transportation industry is no exception, especially due to the impact of the epidemic, rail transit, online car-hailing, bicycle sharing and other new supply travel modes, the traditional "end dish" bus supply service can no longer meet the "door-to-door" and "end-to-end" personalized travel needs of passengers. As a result, bus companies are forced to embark on digital transformation.
In the past few years, the logical starting point of the digital transformation of public transport enterprises is to improve productivity, mainly to implement intelligent scheduling, which solves the problem of low efficiency of traditional manual productivity, but does not solve the problem of scientific decision-making. At this stage, the logical starting point of the digital transformation of public transport enterprises is to meet the changes in passenger demand, and today enterprises are faced with massive, real-time, and multi-scenario passenger demand, so they need to transform themselves into customer (demand) operators to realize the transformation of personalized supply services from "enterprise end dishes" to "passenger à la carte", which not only requires the digitization of management processes, operation and production, but also the resources of the entire enterprise. The decision-making system must be digitized, which requires the construction of a new set of digital capability system to realize the diagnosis, optimization and evaluation of various businesses, so as to establish a business decision-making mechanism based on passenger flow demand, and the base of this digital capability system is the big data platform.
Application scenario of bus big data platform of a bus enterprise).
Difficulties in the construction of big data platforms: data kidnapping and data silos
The biggest difficulty in the construction of big data platform lies in data governance. In recent years, many public transport companies have carried out different degrees of information construction, but due to the lack of unified planning, the construction of information systems separately, resulting in data scattered in the hands of the first businessmen, enterprises seem to have massive data, but they have no authority, no ability to manage and utilize, and then form data fragmentation and islands.
Information system ledger of a public transport enterprise).
For example, before a bus company builds a big data platform, it already has more than 30 information systems such as scheduling, ERP, monitoring, driver behavior analysis, IC card, etc., which include the system developed by the software company, the system presented by the hardware manufacturer, and the self-developed system.
Key points of big data platform construction: data specification and data governance
The primary goal of building a big data platform is to break the data barriers of the original information systems of the enterprise and realize data integration, analysis, sharing and utilization, which requires the establishment of a standard data governance system.
First of all, it is necessary to build a standard data docking interface to facilitate the data access of different information systems and the data access of subsequent expansion systems. Secondly, a standardized data governance mechanism should be established to clean and manage the access of multiple heterogeneous data, unify data standards and norms, break data silos, and lay the foundation for data utilization. Finally, based on the big data framework and big data algorithm model, massive historical data and real-time data are analyzed and calculated, providing data analysis capabilities for upper-layer business display.
The value of big data platform construction: assisting operation and scientific decision-making
The upper-layer business display is based on the platform's big data analysis capabilities, which integrates and analyzes massive data such as people, vehicles, stations, lines, networks, and stations, realizes real-time data feedback and analysis, and plays a role in the whole process of business diagnosis, business optimization, and post-evaluation.
For example, at the level of passenger flow analysis. At present, the biggest pain point of public transport companies is the inability to establish a full range of travel situational awareness, in layman's terms, is the lack of ability to grasp the law of passenger flow. The passenger flow analysis module of the big data platform uses the passenger flow analysis model to diagnose and analyze the passenger flow OD and passenger flow heat of the whole city based on historical bus passenger flow data and real-time bus card data. On a macro level, it can grasp the overall travel rules of the city's bus passenger flow, and provide a basis for the optimization of the line networkAt the micro level, it can understand the changes in passenger flow and passenger retention of a certain line in real time and space, and provide a basis for operation planning and adjustment. Through the analysis and diagnosis of bus passenger flow, we continuously optimize the route network and optimize the operation, so as to reduce the walking time and transfer times of passengers, and comprehensively improve the quality and competitiveness of bus services.
Public transport big data platform - passenger flow analysis: grasp the overall travel rules of urban public transport passenger flow).
Another example is at the level of passenger flow scheduling. At present, the biggest problem faced by bus companies in operation is that they cannot formulate operation plans based on passenger flow rules, which not only leads to long waiting time for passengers, poor travel experience, but also greatly wastes enterprise resources. The passenger flow scheduling application of the big data platform, based on the analysis results of the historical passenger flow data of the bus, comprehensively grasps the rules of the passenger flow of the line in various periods, directions and sections, and then automatically generates the operation scheduling plan through intelligent algorithms according to the passenger flow rules and route parameters (including various constraints and resource allocation conditions such as shift type setting, shift closing mode, energy replenishment parameters, etc.). At the same time, combined with the line portrait, the key indicators such as line revenue, expenses, passenger flow, and full load rate are monitored in real time to realize the monitoring and diagnosis of line operation efficiency, and then continuously adjust and optimize the operation plan according to the diagnosis results.
The passenger flow scheduling application of the big data platform has built the ability for enterprises to save operating costs and increase passenger flow and revenue without sacrificing service or even improving service. Specifically, after a bus company No. 103 adopted passenger flow scheduling, the daily profit still increased by an average of 632 yuan in the case of reducing 1 driver, 1 car and 12 trips per day.
Bus Big Data Platform - Passenger Flow Scheduling: Automatically generate an operation plan according to the rules of different passenger flows on the route).
Bus Big Data Platform - Line Portrait: Real-time Monitoring and Diagnosis of Route Operation).
Another example is at the level of security analysis. At present, the biggest cause of bus safety accidents is the driver's human factor, and the most headache for bus companies is also the management of drivers. The safety analysis module of the big data platform constructs a driver's portrait by analyzing the driver's basic information data, safety behavior data, operation and production data, etc., and conducts a comprehensive score according to its 100 kilometers income, planned trip completion rate, comprehensive punctuality rate, violations, praise times and other indicators. Exams to enhance drivers' safety awareness and service awareness.
The security analysis application of the big data platform not only improves the efficiency of enterprise security supervision, but also greatly reduces the security cost of enterprises. Specifically, after using the platform, the cost of an accident per 100 kilometers was reduced from 20 yuan to about 17 yuan per 100 kilometers, and at the same time, 18 safety administrators were optimized, saving about 1.5 million yuan in labor costs throughout the year.
Bus Big Data Platform - Driver Portrait: Real-time monitoring and diagnosis of driver behavior and operation).
Another example is the operational analysis level. At present, the public transport industry can not get full subsidies, is not entirely a problem of financial funds, for the city with financial capacity, the phenomenon of empty bus and low load, so that the investment funds and income are not proportional, so that the first department feels that the financial investment is a waste. Therefore, it is very important to improve the efficiency of public transport resource investment. The operation analysis module of the big data platform establishes a real-time monitoring mechanism for key indicators such as 100 kilometers of income, 100 kilometers of passenger flow, 100 kilometers of power consumption, and 100 kilometers of accident cost through the integration and analysis of various operating data such as enterprise income, passenger flow, power consumption, violation of regulations, alarms, and arrival rate, so as to realize cost-benefit analysis and resource investment efficiency analysis, provide a basis for cost regulation, cost reduction and efficiency increase of public transport enterprises, and also provide a basis for measuring the actual effect of financial subsidies.
Bus Big Data Platform - Operation Analysis: Real-time monitoring and diagnosis of the overall operation of the enterprise).
Summary
Based on the self-developed data platform, the public transport big data platform built by Smart Changxing takes the governance, analysis, sharing and utilization of public transport data as the core, through the fusion and analysis of data such as people, vehicles, stations, lines, networks, and stations, and combines the self-developed bus model algorithm to profile drivers, vehicles, stations, routes, and enterprises, so as to realize the diagnosis of various businesses (find out business problems and gaps), optimize (give business optimization suggestions) and evaluate (evaluate the optimization results) In order to help enterprises establish a business decision-making mechanism based on passenger flow demand, realize data-assisted operation and production and scientific decision-making, and comprehensively improve the efficiency of public transport resource investment, and promote the scientific development of digital and intelligent public transport.
Author: Zhao Congyun, an expert in the public transportation industry and general manager of Smart Changxing.
Company Profile:
Founded in 2015, Hunan Smart Changxing Transportation Technology Co., Ltd. is a data-driven scientific and technological innovation enterprise of Internet + transportation, the company takes vehicles as a link, based on advanced technologies such as the Internet of Things, big data analysis, artificial intelligence, cloud computing, and intelligent hardware, and provides advanced technologies for the first department (urban traffic brain), transportation enterprises (smart bus cloud platform), and passengers (MaaS travel services) Provide one-stop digital intelligence solutions to comprehensively help transportation enterprises improve the construction and operation efficiency of information facilities, improve the utilization rate of social vehicle resources and the efficiency of resource allocation, meet the needs of the public for personalized travel, and help the transportation industry fully realize digital and intelligent transformation.