Despite such strategies already in place, what went wrong on Mauni Amavasya, January 29, when the unfortunate stampede happened claiming so many lives? What were the major lapses?It was extremely unfortunate because we had been working to avert any such tragedy for over the last two years. Before January 29, there were a lot of positive stories about Mahakumbh which were the result of our hard work of two years. However, the unfortunate and painful incident was all because of the massive crowd pressure. However, on this particular question there’s a judicial commission which is looking into it. So, I would not like to share my views because I have to give my views before the judicial commission.After the Mauni Amavasya stampede, what were the special measures taken by the administration for succeeding snans so as to avert the recurrence of a mishap?There was a review meeting convened by the state Chief Secretary and Director General of Police immediately on January 30. The joint team of Chief Secretary and the DGP visited the mela and the areas where the possible weaknesses had developed and addressed them. It was followed by a high-level meeting by the Chief Minister on the same day itself. The CM also visited the mishap site on the Mela premises. After the visit of the CM, the Chief Secretary and the DGP, we reviewed the areas where we were doing better and also those areas where we needed to improve. We picked up from where we had left and started preparing for the next snan. And then all the CCTV cameras were analyzed and the strong points were enhanced and the gaps were filled up because there was a huge crowd surge that was coming up. So all the gaps were re-looked into and additional police and administrative personnel were posted to keep a close watch of the crowd to regulate its movement.What has been the role of Artificial Intelligence (AI) as it was a highly-publicized tool which was being used for the first time for crowd management and monitoring other aspects related to the mela?It was very, very useful. Basically, Artificial Intelligence was used for the first time in the world for crowd management because the existing software was not aware of how to give input. So, we had been machine learning these softwares for over six months and testing them on Dussehra, Diwali and other such occasions when the crowd surges in respectable numbers at various specific places. Using all those inputs, we machine-learned the existing software so as to get credible inputs. The inputs thus received were very, very useful over a period of time. They could give us the density mapping and other parameters including the rate at which the crowd was surging.
Source link
