طراحي سامانه مبتني بر اطلاعات جغرافيايي داوطلبانه جهت ارزيابي سريع خسارت زمين‌لرزه

Designing a Volunteer Geographic Information-based service for rapid earth quake damages estimation

براي مديريت بحران در زمان وقوع زلزله و كاهش آسيب­ هاي ناشي از آن، حجم گسترده ­اي از اطلاعات در زمان­ هاي اوليه پس از وقوع زلزله ضروري است. مشكل اصلي سيستم‌هاي موجود براي برآورد خسارت اين است كه اطلاعات را نمي­ توانند به صورت لحظه اي در هنگام وقوع فاجعه مخابره كنند و بر اساس اطلاعات از پيش جمع‌آوري‌شده اقدام به برآورد مي شود. در سامانه طراحي شده تحت وب در اين تحقيق, كاربران با به اشتراك‌گذاري لحظه اي داده هاي مربوط به خسارت واردشده به خود يا ساير افراد، حجم گسترده­ اي از اطلاعات را براي تحليل در اختيار تيم مديريت بحران قرار مي ­دهند. سيستم توسعه داده‌شده علاوه بر جمع آوري و ذخيره اطلاعات داوطلبانه, آناليز مكاني Heatmap را براي بررسي پراكنش مكاني و نمايش خسارت انجام مي دهد. براي بررسي نتايج, سيستم به ­صورت فرضي در شهر اشنويه در غرب استان آذربايجان غربي پياده ­سازي شده ويك سناريوي فرضي براي زلزله طراحي شد. پس از به اشتراك‌گذاري اطلاعات توسط مردم داوطلب، Heatmap ميزان خسارت در زمان كوتاهي توليد و در اختيار مديران بحران كه در اجراي اين طرح همكاري كردند قرار گرفت تا درك مناسبي براي تصميم­ گيري در هنگام بروز زلزله ­هاي احتمالي به دست آيد. نتايج نشان مي دهد كه پياده ­سازي اين سيستم­ علاوه بر كاهش چشمگير سرعت جمع ­آوري اطلاعات، كاهش زمان تحليل اطلاعات بر اساس توليد Heatmap را به همراه خواهد داشت به طوري كه استفاده از اطلاعات جغرافيايي داوطلبانه باعث افزايش 6.5 برابري سرعت و زمان تخمين خسارت در مقايسه با روش هاي سنتي موجود مي شود به طوريكه مي تواند به عنوان يك روش و ديدگاه نوين در مديريت مخاطرات محيطي مورد استفاده قرار گيرد.

Introduction The advent of Web 2.0 enables the users to interact and prepare free unlimited real time data. This advantage leads us to exploit Volunteer Geographic Information (VGI) for real time crisis management. Traditional estimation methods for earthquake damages are expensive and time consuming. In contrast, volunteer and web-based service are near real time with almost no cost services. the lack of accessible real time data collection services causes delayed-emergency responses for disasters like an earthquake. This drawback is critical when we encounter a problem like buried people with valuable seconds for emergency rescue operation. The current research aims to design and implement a web-based volunteer data collection service for rapid estimation of earthquake damages and number of buried people. Methodology To investigate the capacity of VGI in rapid estimation of earthquake, a technical frame work based on the web technologies has been programmed and implemented. The designed service is comprised of server and client sides. The client side is a two-side browser-based service includes volunteers (users) and managers pages. On the user page, volunteers have a web page to enter and fill in the blank forms and taking a photograph of the target building and compare it with pictures. They watch the sample pictures in different level of damages and compare their building with the samples and give a grade of the most similar picture with their building. This grading process leads the server to analysis and classify the incoming data and create the heatmaps for managers. On the managers page two online discrete heatmaps for the both earthquake damages and buried people are displayed. In fact, the heatmaps present the online and real time quantitative situation of the building damages and buried persons as hot spots. These hotspots have the first priority for giving emergency services. The manager page also exploits query tools to request different level of details and classes from the server side. The server side is responsible for receiving, saving, spatial analysis and transmission of the requested result to the client side. This task is carried out by the exchange side. As the citizens are entered to the browser-based service and fill in the blank forms for building damages based on the mentioned guideline and report the buried people, These forms are transmitted to the server side and a geo-server performs spatial analysis including Heatmap, distance and clustering analysis. Then, a real time damage and buried people map are prepared and delivered to the client side. The server updates the created maps whenever a new data is submitted. By this, a real time damage and buried people maps are accessible for official managers to conduct a goal-oriented emergency operation and a preliminary earth quake damages on city building blocks. After the technical frame work has been designed, it was tested in Oshanvieh city by 132 volunteers on the scene for an earthquake. Results and discussion To investigate the capability of volunteer geographic information for earth quake afterwards, the designed service mentioned in the methodology was utilized on Oshnavieh city. It was assumed that an earthquake has occurred. 132 volunteers participated for the data collection process. According to the crisis management organization experts, 102 reports of the total 132 reports are correct that shows the accuracy of 76.52 percent. Besides the building damage level based on the defined guideline, the citizens also select their vital needs after the earthquake. the most requested vital needs are warm stuffs, medicine, water and foods respectively. Unfortunately, the participation rate is ranged from some seconds after the earthquake to three days. This means that some citizens have filled and transmitted their data three days after the earthquake. In the following, a comparison between the designed service and traditional earthquake damage estimation methods (in situ) was carried out. The result shows that field-based methods for a city like Oshnavieh need about 20 days. However, the designed volunteer-based service what is programmed and implemented in the current research does this job by 3 days. Conclusion As the results show, the proposed service designed in this research implements the damage estimation process 6.5 times faster than the governmental procedures. This proves the efficiency of the research achievements. Besides the velocity, traditional damage estimation methods are expensive compare to volunteer-based data collection and processing which are almost free, scalable and pervasive.