Skip to main content
QUICK REVIEW

[论文解读] All you can stream: Investigating the role of user behavior for greenhouse gas intensity of video streaming

Paul Suski, Johanna Pohl|arXiv (Cornell University)|Jun 19, 2020
Green IT and Sustainability参考文献 27被引用 4
一句话总结

本研究通过结合生命周期评估(LCA)与对91名用户的7天在线调查,探究用户行为如何影响视频流媒体的温室气体(GHG)强度。研究发现,使用智能手机流媒体每小时排放的二氧化碳最多是使用智能电视的10倍,且默认低分辨率设置可显著降低用户端的能源需求,凸显了行为与设计在实现可持续流媒体方面的关键作用。

ABSTRACT

The information and communication technology sector reportedly has a relevant impact on the environment. Within this sector, video streaming has been identified as a major driver of CO2-emissions. To make streaming more sustainable, environmentally relevant factors must be identified on both the user and the provider side. Hence, environmental assessments, like life cycle assessments (LCA), need to broaden their perspective from a mere technological to one that includes user decisions and behavior. However, quantitative data on user behavior (e.g. streaming duration, choice of end device and resolution) are often lacking or difficult to integrate in LCA. Additionally, identifying relevant determinants of user behavior, such as the design of streaming platforms or user motivations, may help to design streaming services that keep environmental impact at a passable level. In order to carry out assessments in such a way, interdisciplinary collaboration is necessary. Therefore, this exploratory study combined LCA with an online survey (N= 91, 7 consecutive days of assessment). Based on this dataset the use phase of online video streaming was modeled. Additionally, factors such as sociodemographic, motivational and contextual determinants were measured. Results show that CO2-intensity of video streaming depends on several factors. It is shown that for climate intensity there is a factor 10 between choosing a smart TV and smartphone for video streaming. Furthermore, results show that some factors can be tackled from provider side to reduce overall energy demand at the user side; one of which is setting a low resolution as default.

研究动机与目标

  • 识别显著影响视频流媒体温室气体(GHG)强度的用户行为因素。
  • 评估社会人口统计、动机和情境因素如何塑造流媒体选择及其环境影响。
  • 探讨平台设计(如默认分辨率设置)如何减少用户端的能源消耗。
  • 将用户行为数据整合到生命周期评估(LCA)模型中,以实现更准确的环境影响评估。

提出的方法

  • 开展为期7天的在线调查(N=91),收集用户流媒体行为的实时数据,包括设备类型、分辨率、时长和使用情境。
  • 将调查数据与生命周期评估(LCA)方法相结合,建立视频流媒体使用阶段的排放模型。
  • 通过为每类流媒体会话分配基于设备和分辨率的二氧化碳排放系数,将用户行为映射到环境影响。
  • 分析社会人口统计、动机和情境变量作为流媒体行为潜在决定因素的影响。
  • 采用统计建模方法识别显著影响GHG强度的关键行为杠杆。
  • 提出设计干预措施,如将低分辨率设为默认设置,以降低整体能源需求。

实验结果

研究问题

  • RQ1使用终端设备(如智能手机与智能电视)的选择如何影响视频流媒体的二氧化碳强度?
  • RQ2用户对分辨率和流媒体时长的偏好在多大程度上影响视频流媒体的环境影响?
  • RQ3哪些社会人口统计、动机或情境因素显著预测更高的或更低的流媒体相关排放?
  • RQ4平台设计选择(如默认分辨率设置)能否被用于减少用户端的能源消耗和排放?
  • RQ5如何通过整合真实用户行为数据来增强生命周期评估(LCA)模型?

主要发现

  • 视频流媒体的二氧化碳强度因所用终端设备不同而相差达10倍,使用智能手机时每小时排放的二氧化碳显著高于使用智能电视。
  • 由于能效和屏幕尺寸的差异,使用智能手机流媒体的碳排放量最高可达使用智能电视的10倍。
  • 在流媒体平台上将低分辨率设为默认设置,可显著降低用户端的能源需求和总体排放。
  • 用户行为受情境因素(如地点、一天中的时间、设备可及性)的强烈影响,这些因素也影响环境影响。
  • 年龄和教育水平等社会人口统计因素与流媒体习惯及关联排放存在相关性。
  • 将真实用户行为数据整合到LCA模型中,与仅基于技术的评估相比,揭示了显著差异,凸显了在环境评估中纳入行为因素的必要性。

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。