Pollution Avoidance Support System (PASS) using GIS, Machine Learning and Big Data

空气污染每年造成数百万人死亡。就像“慢镜头大流行”一样，肮脏的空气是我们健康的瘟疫，每年在全世界造成700万人死亡和许多可预防的疾病，如中风、心脏病、肺癌和急性呼吸道感染（世卫组织，2021年）。每年，在英国，它导致36,000人过早死亡（政府空气污染物医学影响委员会comap），并造成200亿英镑的损失。尽管大流行引发的封城导致2020年全球年度排放量下降幅度最大，但放宽封城措施的幅度已经超过了大流行前的水平。尽管许多政府采取了各种行动（例如制定清洁空气区），但预计恶劣的空气质量将持续到2050年（经合组织，2019年）。科学家建议避开易受污染影响的人群，如患有呼吸系统疾病（如哮喘、支气管炎等）的人，他们会出现并发症，有时会因暴露于高污染水平而死亡（欧洲公共卫生联盟，2020年）。污染水平在一个城市/城镇的许多地点之间差异很大，并且在一个地点的不同时间会有所不同。目前的解决方案提供了全市范围的信息，因此对于躲避方法无效，因为弱势群体无法分辨出在城市/城镇中应该使用或避免哪些位置，如果他们外出散步/旅行/锻炼等。采用规模小得多的污染数据的解决方案，例如邮政编码单位级别，可以解决这个问题。然而，英国没有，也可能不可能为其大约170万个邮政编码单位（一个城市/城镇可能有100个邮政编码单位）中的每一个都配备监控设备。尽管如此，在英国各地运行的数千个排放传感器（BBC, 2019）提供了足够的数据，如果部署了正确的工具，可以为所有邮政编码单元开发模型。因此，该项目旨在开发一个系统，通过机器学习算法、地理信息系统数据、远程信息处理、天气数据和大数据分析，为用户提供邮政编码单位特定的污染数据。该系统将通过网络和移动应用程序提供，并将包括** live - pass:**将为每个英国邮政编码单元提供实时污染数据，以支持用户决定是否在特定地点/邮政编码单元进行户外活动（例如旅行，户外运动等）。它将提出更清洁的替代方案。** future - pass:**将为每个英国邮政编码单位提供每小时7天的未来污染预测，以支持规划/安排未来户外活动，选择更清洁的地点/时间。**城市/城镇分析仪表板（CAD）：**提供城市/城镇内所有邮政编码单位的不同污染水平的数据和见解（针对地方当局）

Air-pollution kills millions every year. Like a 'pandemic in slow motion', dirty air is a plague on our health, causing 7-million deaths and many preventable illnesses like stroke, heart disease, lung cancer and acute respiratory infections worldwide each year (WHO, 2021). Yearly, in the UK, it causes 36,000 premature deaths (Government's Committee on Medical Effects of Air-Pollutants COMEAP) and costs £20-billion.Although, pandemic-induced lockdowns caused largest drop in annual global emissions in 2020, lockdown easing has seen a surge to more than pre-pandemic levels. Despite various actions taken by many governments (e.g. enacting clean-air-zones), poor air quality is projected to continue into 2050 (OECD,2019).Scientists recommend dodging approach to pollution-vulnerable people like those with respiratory illness (e.g. asthma, bronchitis, etc.) who develop complications and sometimes die due to exposure to high pollution levels (European Public Health Alliance, 2020). Pollution levels can vary widely between many locations within a city/town and will vary from time to time for a location. Current solutions provide city-wide information and are thus ineffective for dodging approach as a vulnerable person is unable to decipher which location(s) within a city/town to use or avoid if they were out on walk/journey/exercise etc. A solution with pollution-data on a much smaller scale, e.g. at postcode-units level, can solve this problem. However, the UK does not, and probably cannot, have monitoring equipment for each of its approximately 1.7 million postcode-units (a city/town can have 100s of postcode-units). Nonetheless, the thousands of emission-sensors operational across UK (BBC, 2019) provide enough data to develop models for all postcode-units if the right tools are deployed.This project therefore aims to develop a system that can provide postcode-units-specific pollution data to users using machine learning algorithms, GIS data, telematics, weather data and big data analytics. The system will be available via web and mobile app and will include**Live-Pass:** will provide live pollution data for each UK postcode-unit to support users in deciding for or against an outdoor activity (e.g. journey, outdoor exercise etc.) in a specific location/postcode-unit. It will suggest cleaner alternatives.**Future-Pass:** will provide hourly 7-day future pollution forecast for each UK postcode-units to support planning/scheduling future outdoor events for a cleaner location/time.**City/town analytics dashboard (CAD):** provide data and insights on the different levels of pollution for all the postcode-units within a city/town (targeted at local authorities)