出租车新运营模式下的LED广告精准投放策略

doi:10.16381/j.cnki.issn1003-207x.2020.10.022

摘要/Abstract

摘要： 出租车LED广告是出租车广告的重要形式，其广告效果与投放策略息息相关。出租车在传统运营模式下，出租车广告投放策略比较粗放，没有考虑消费者的差异化需求。随着打车软件应用的普及，出租车运营模式发生了根本性改变，特别是在新的运营模式下积累了大量的运行数据，为出租车广告投放策略的精细化设计和精准投放提供了必要条件。本文针对新的运营模式下的广告精准投放问题进行研究，构建了提取居民出行行为时间特征和空间特征以及出租车司机行为特征的方法，并针对居民出行行为规律和司机的行为模式和偏好，提出了出租车LED广告精准投放的三种策略，即时段定向投放策略，空间定向投放策略和时空定向投放策略。最后，利用某市实际的出租车轨迹数据进行实例分析。本文提出的出租车广告精准投放策略的设计思路和方法对于相关研究具有一定参考价值，提出的精准投放策略对出租车广告投放和管理相关部门的决策也具有实际参考价值。

关键词: 出租车广告, 精准投放, 打的软件, 司机行为, 出行行为

Abstract: As a new form of taxi advertising, taxi LED advertisement is deeply concerned and favored by advertisers, and its advertising effect is closely related to the delivery strategy.The traditional operational mode of taxi is mainly single shift system and double shift system.In the traditional operational mode, the LED advertising time is not flexible enough, the content is not rich enough, and the delivery area is not fine enough.In short, the current taxi LED advertising strategy is relatively extensive, and does not take into the differentiated needs of target consumer groups account.
With the popularity of ride-hailing software applications, people have experienced the convenience of the Internet to travel.The taxi industry is experimenting with a new model of "Internet travel", suggesting that taxi operations are fundamentally changing.Especially in the new operation mode, the taxi equipped with GPS positioning equipment has accumulated a large amount of running data during the driving process, which provides the necessary conditions for the detailed design and accurate delivery of the taxi LED advertising strategy.
The current theory and application of precision advertising research are analyzed, aiming at the new operation mode of taxi LED advertising is not accurate. A method to extract the temporal and spatial characteristics of resident travel behavior and the behavior characteristics of taxi drivers is constructed. According to the behavior patterns and preferences of residents and drivers, three precise strategies of taxi LED advertising are put forward.Immediate segment oriented delivery strategy, space oriented delivery strategy and time-space oriented delivery strategy.Finally, the real track data of a city taxi is used to verify and analyze it.The specific contents of the study are as follows:
(1) The feature extraction of residents' travel behavior.In this paper, the difference function algorithm is used to find the peak and off-peak travel time, and a method to extract the travel behavior time feature is proposed.Secondly, the peak density clustering algorithm is used to cluster the spatial longitude and latitude data, and the map matching and regional functional identification are carried out to get the spatial characteristics of residents' travel.
(2) The feature extraction of drivers' behavior.Firstly, the preference degree of different drivers from different drivers to different functional regions is calculated under the condition of not carrying passengers, and then the K-Means clustering algorithm is used to cluster the driver preference.Finally, based on the clustering results, the characteristics of drivers with similar behavior and those with different behaviors were observed and analyzed respectively.
(3) The formulation of precision delivery strategy.Based on the behavior rules of residents and the behavior habits of drivers, this paper puts forward three kinds of directed delivery strategies, namely, immediate segment orientation, spatial orientation and spatio-temporal orientation.Because the travel behavior of residents fluctuates in time, according to the change of residents' behavior with time, according to the change of residents' behavior with time, the timely advertisement is put out, and the appropriate advertising content is published.Set up reasonable advertising frequency and develop differentiated advertising price.The spatial orientation strategy is to place advertisements on the functional areas that residents often travel to and from.The longitude and latitude data reflecting spatial location are classified into urban functional areas by clustering, map matching and regional functional identification, and then the types of advertising information are determined in which functional areas.Spatiotemporal targeting strategy refers to placing advertisements in different functional regions at different times.Spatio-temporal directed delivery needs to mine the behavior characteristics of taxi drivers in different time and different functional areas, find the drivers with similar characteristics, and extract the functional areas of interest to achieve accurate advertising.
The design ideas and methods of the precise placement strategy of taxi LED advertisement are pect forward, improves the theoretical research system of taxi advertisement is improued, and the field of precision advertising is widened.The proposed precise delivery strategy has practical reference value for advertisers, advertising companies, taxi companies and other management departments involved in taxi advertising.

Key words: taxi advertisement, accurate delivery, taxi-hailing applications, drivers' behavior, travel behavior

中图分类号:

C934

李科, 党延忠. 出租车新运营模式下的LED广告精准投放策略[J]. 中国管理科学, 2020, 28(10): 220-230.

LI Ke, DANG Yan-zhong. The Precise Delivery Strategy of LED Advertisement in New Operational Mode of Taxi[J]. Chinese Journal of Management Science, 2020, 28(10): 220-230.

参考文献

[1] 杨洋. "打的软件"对出租车司机行为决策及运营绩效的影响研究——技术效率与配置效率的双重视角[D]. 杭州:浙江大学, 2016.
[2] 李郁. 出租车行业"互联网+"运营模式变革探讨[D]. 昆明:云南财经大学, 2016.
[3] Castro P S, Zhang D, Chen C, et al. From taxi GPS traces to social and community dynamics:A survey[J]. Acm Computing Surveys, 2013, 46(2):1-34.
[4] 牛森. 基于用户兴趣的IPTV广告精准投放算法[J]. 计算机应用与软件, 2017, 34(6):221-225.
[5] Wang C, Zhang P, Choi R, et al. Understanding consumers attitude toward advertising[C]//Eighth Americas Conference on Information Systems. 2002:1143-1148.
[6] Grossman G M, Shapiro C. Informative advertising with differentiated products[J]. Review of Economic Studies, 1984, 51(1):63-81.
[7] Chatterjee P, Hoffman D L, Novak T P. Modeling the clickstream:implications for Web-Based advertising efforts[J]. Marketing Science, 2003, 22(4):520-541.
[8] Athey S, Ellison G. Position auctions with consumer search[J]. Quarterly Journal of Economics, 2011, 126(3):1213-1270.
[9] 邓瑛. 精准广告的传播模式与策略研究[D]. 长沙:湖南大学, 2008.
[10] 鞠宏磊, 黄琦翔, 王宇婷. 大数据精准广告的产业重构效应研究[J]. 新闻与传播研究, 2015,(8):98-106.
[11] 倪宁, 金韶. 大数据时代的精准广告及其传播策略——基于场域理论视角[J]. 现代传播-中国传媒大学学报, 2014, 36(2):99-104.
[12] Ha I, Oh K J, Jo G S. Personalized advertisement system using social relationship based user modeling[J]. Multimedia Tools & Applications, 2015, 74(20):8801-8819.
[13] Broder A, Fontoura M, Josifovski V, et al. A semantic approach to contextual advertising[C]//The 30th ACM SIGIR conference on research and development in information retrieval, Amsterdam, July 23-27,2007.
[14] Lim J, Kim M, Lee B, et al. A target advertisement system based on TV viewer's profile reasoning[J]. Multimedia Tools & Applications, 2008, 36(1-2):11-35.
[15] 齐观德. 基于大规模出租车轨迹的移动感知计算研究[D]. 杭州:浙江大学, 2013.
[16] 童晓君, 向南平, 朱定局. 基于出租车GPS数据的城市居民出行行为分析[J]. 电脑与电信, 2012(z1):56-59.
[17] 李艳红, 袁振洲, 谢海红,等. 基于出租车OD数据的出租车出行特征分析[J]. 交通运输系统工程与信息, 2007, 7(5):85-89.
[18] 梁军辉, 林坚, 杜洋. 大数据条件下城市用地类型辨识研究——基于出租车GPS数据的动态感知[J]. 上海国土资源, 2016, 37(1):28-32.
[19] 陈世莉, 陶海燕, 李旭亮,等. 基于潜在语义信息的城市功能区识别——广州市浮动车GPS时空数据挖掘[J]. 地理学报, 2016, 71(3):471-483.
[20] 韩吉德, 王祖顺, 王霞. 基于出租车轨迹数据的人口活动分析[J]. 测绘工程, 2016, 25(10):69-71.
[21] 金鹏. 基于用户行为的网络广告精准投放平台的研究[D]. 武汉:华中师范大学, 2015.
[22] Brahim E B, Lahmandi-Ayed R, Laussel D. Is targeted advertising always beneficial?[M]//Hadri K,Mikhail. Econometric Methods and Their Applications in Finance, Macro and Related Fields,2014:551-585.
[23] 赵江. 基于电商平台的定向广告投放机制和策略研究[D]. 南京:东南大学, 2015.
[24] Rodriguez A, Laio A. Machine learning. Clustering by fast search and find of density peaks[J]. Science, 2014, 344(6191):1492-1496.
[25] 黄岚, 李玉, 王贵参,等. 基于点距离和密度峰值聚类的社区发现方法[J]. 吉林大学学报(工), 2016, 46(6):2042-2051.
[26] 袁军辉. 基于多特征信息融合的WEB广告分类方法研究[D]. 哈尔滨:哈尔滨工业大学, 2013.