Hidden in Plane SightFinding Lost Airplanes Before Time Runs Out-2015美赛原版-L文.pdfVIP

Hidden in Plane Sight: Finding Lost Airplanes Before Time Runs Out When a plane disappears over remote waters, time is of the essence. The faster we learn the fate of the plane, the more likely it is that we can rescue survivors or bring closure to grieving families. In light of the failed search effort for Malaysia Flight MH370, we must strive to expedite and optimize response effectiveness in the future. First we investigate how and where airplanes crash and discover that finding a lost plane begins with predicting its location based on variables such as flight speed, glide ratio, and altitude. We also consider the search plane’s speed and detection distance. Our simulations test different flight paths to determine which best minimizes search time and maximizes chances of finding the crash site debris. ● Where to look: ​Projecting a commercial airplane’s flight path, we construct a Gaussian normal distribution bounded by a rectangular search area. ● Simulate Solutions with Ideal Accuracy: ​Assuming the search will certainly detect a wreck within radar range,​ ​we test random flight paths, but they prove inefficient—the most successful takes 38 hours. By searching systematically, reducing redundancy and focusing on high probability areas, we cut mean search time to 3.6 hours. ● Simulate Solutions with Imperfect Accuracy:​ We now incorporate uncertain detection probability into our model. Applying Bayesian Search Theory to our most efficient solution results in a relatively low mean search time of 4.7 hours, with 95% of crashes located within one day. ● Flexibility: ​We can easily repurpose our solutions for different numbers of search planes or sizes of crashed planes. Ultimately, our recommended solution — an application of