Statistics for Moscow in the scenario “people try to sit at home, there is no air traffic” - by November, the model shows 5 million patients. This is a limited forecast based on incomplete data, below are the details. Set to zero on March 22.Several models of the spread of infection were created in the world, but none of them was suitable for Russia, or relied on population density without the right graph of movements of people. Why? Because either it turns out to be so difficult that you are okay to coordinate it, or simply no one in this place has this dataset.Except us.Tutu.ru has been happy to share data with journalists for 16 years (a huge part of the news in the spirit of “Anomalous demand for Antalya is noticeable” is a cut of our information windows). But we have never historically disclosed the data on the movements of people in whole blocks.We collected a dataset of people moving around Russia in April 2019 and transferred it to the Open Data Science community . If you don’t know them, this is an association of predominantly Russian data-scientists (but from all over the world) that processes open data into utility models. Non-profit.Below are the conclusions, a table with a forecast for each large city, the dataset itself (if you want to try to do something with it). About how the model works and what mathematics and limitations are inside, ODS will tell you in a couple of hours. And lay out the source. UPD: here .Datacet
This is how people traveled around the country over the past April-2019 (with some errors). The dataset is a set of vectors from city to city (the first city indicated is where, the second is where), the type of transport and the number of passengers restored to 100%. A dataset is anonymous statistical data relating to groups of people.Here is the visualization of the set . Thanks again for herSafronov.Data limitations : buses are the most inaccurate part of the dataset. We cannot know exactly how many people traveled by bus because of the so-called “gray” carriers, which we do not support on the platform. But we tried to restore this data on known routes. In the air and railway data are much more accurate, but not 100%. We do not see the movement of the military, railway personnel, children's cars and other unusual tickets. There are a number of shipments such as helicopter routes between the cities of the Far East and the propulsion aircraft of Yakutia. In aviation, our market coverage is very good throughout the European part of Russia and falls to the east (in Vladivostok, Novosibirsk and Khabarovsk, the data are most accurate in the eastern part of the country). On train tickets, the error is quite small.If a person was traveling on a Moscow-Petersburg train and left in Tver, then he is considered to be a Moscow-Tver passenger.More accurate data can be obtained through different carriers, departments and official statistics, but this is almost impossible in practice in a short time. Our data are sufficient for evaluation, but just remember that they are also collected and restored with some error.You can pick it up here . In a couple of hours, there will be a publication about how the model works and what is under its hood, and ODS will open the source. There will be a repository with this dataset already installed and others (like a matching city map by name with coordinates and the number of cases).If you do something with it, please show me in a personal email or in my mail noreply@tutu.ru.Scenarios
There are three basic scenarios: “we don’t touch anything, everything goes as usual” - then everything is predictably bad in terms of infection. The second scenario is “we block all air traffic and people try to stay at home, but cars and trains continue to ride.” The third scenario is the closure of the main transport hubs. Taking into account recent changes, we calculated the first one (there are conclusions in CSV), but also took the conclusions of the fourth, where we modeled the reduction of traffic in the country to 10% of the usual.Follow the same link for CSV scripting.This is not the first time the world community has encountered epidemics and is not the first time has predicted their development with the help of matmodels. The mathematics is complicated in places, neural networks are not needed. But modeling flows between vertices of a graph is mathematically very closeto the most modern architecture of neural networks. The epidemic spreading algorithms have long been known, you just need to set parameters like contagiousness. Which were calculated for us by the Chinese, Italians and others who had encountered the problem earlier. The last post with a bunch of links to research, verified by doctors, was largely a collection of initial data for the model. Nevertheless, I warn you again: the model uses not very accurate initial data, there are no professional epidemiologists among the developers (but we use their algorithms), the model has its limitations. Details on SIR will be in the ODS post. Estimated accuracy - up to order.The model’s work looks like this:- We consider the spread of the disease per day.
- We count how many people moved to another city based on weighted transportation vectors.
- We recount the number of infected in cities.
- We start the next measure.
The noise at the beginning of the model is caused by the fact that the starting point is March 22, 2020, and it does not take into account those who got infected somewhere abroad before and did not manifest themselves before the test in the following days. It is also important that the reference data of the model is not the actual number of patients, but the number tested with a positive test for COVID-19. That is, there may be more carriers in fact, and the infection cycle will be reduced. Infections inside the vehicle are not yet taken into account in the model.results
I show two calculated extremes - what will happen if you do nothing (option 1) and take a maximum of measures, but do not turn on the total quarantine mode with quarterly boundaries (option 2).All on the same CSV download link , in the table format below.Scenario 1: the fastest
Scenario 1 is the worst from the point of view of the spread of infection, when 100% of traffic between cities remains (now it is already lower), and people do not try to isolate themselves, for example, ride buses and the metro to such fairs , but at the same time they comply with recommendations for washing hands and try to maintain a distance (with varying success). It is simulated for six months, therefore, for example, Moscow will not be cured to a state of “less than a thousand infected at the same time” as part of the model’s development period.Column parameters - the number of patients infected at the same time (the survivors are not included). The first threshold begins with "more than a thousand" (this is the day when the number of infected people in the city exceeds 1,000), then 10 and 100 thousand. The fourth column is the moment when subjectively you can stop hiding, less than 1000 at the same time infected. The numbers in the columns are the day the threshold is reached. For example, Moscow is gaining the first thousand concurrently infected models predicted in this scenario in 13 days. >1000 >10.000 >100.000 <1000
13 25 38 -
- 23 33 45 -
28 40 57 -
30 41 55 -
-- 30 41 56 -
32 44 - 172
32 44 64 174
33 44 58 -
33 45 61 -
34 47 - 171
35 47 64 -
35 46 63 -
35 46 64 -
() 36 48 - 178
37 49 - 177
37 49 - -
37 48 63 -
37 48 66 -
37 48 63 -
37 49 - 175
38 49 67 -
38 48 62 -
38 48 63 -
38 50 - -
38 50 - -
38 50 - 171
39 50 - 176
39 51 - 174
40 54 - 152
- 40 54 - 149
40 52 - 165
40 52 73 -
41 59 - 140
41 53 - 167
41 53 - 165
42 54 - 160
42 53 - 174
42 54 71 -
42 54 - 170
42 53 - -
42 59 - 148
42 54 - 174
42 60 - 140
42 53 69 -
42 54 - 169
43 55 - -
43 63 - 133
43 55 73 -
43 54 70 -
43 55 - 168
44 57 - 149
44 62 - 134
- 44 68 - 131
44 55 72 -
44 61 - 136
45 66 - 132
- 45 57 - 171
45 67 - 130
45 58 - 161
45 57 - 164
45 56 - -
45 61 - 138
46 62 - 136
46 - - 126
46 - - 124
46 - - 127
46 58 - 176
46 59 - 152
46 57 75 -
46 58 - 174
46 - - 123
46 58 80 -
46 58 75 -
47 64 - 143
47 - - 124
47 61 - 146
47 61 - 156
47 61 - 146
47 60 81 -
47 59 77 -
47 59 81 -
47 59 - 177
47 61 - 151
47 60 - -
47 - - 126
47 58 78 -
47 61 - 154
47 60 - 169
48 61 - 176
48 62 - 145
48 60 - -
48 - - 123
48 - - 115
48 - - 122
48 60 - 158
48 62 - 152
48 61 - 150
48 - - 115
48 67 - 131
48 64 - 135
48 66 - 131
48 65 - 132
49 - - 126
49 71 - 129
49 63 - 147
49 - - 122
49 - - 118
49 62 - 159
49 - - 115
49 - - 129
49 - - 127
49 63 - 149
49 - - 123
49 69 - 130
49 64 - 149
49 - - 118
49 62 - -
49 - - 116
49 61 - -
49 62 - 151
50 - - 117
50 65 - 142
50 65 - 151
50 63 - 160
50 - - 119
50 - - 124
50 - - 127
50 - - 115
50 - - 130
50 64 - 150
50 66 - 137
50 65 - 147
51 64 - 156
51 65 - 154
51 64 - 170
51 66 - 140
51 64 94 -
51 65 - 160
51 65 - 163
51 64 - -
51 - - 115
51 - - 118
51 - - 115
51 64 - 165
51 67 - 140
51 - - 122
51 67 - 140
51 65 - 160
51 - - 123
51 63 84 -
51 66 - 143
51 64 92 -
- 51 - - 113
52 - - 128
52 - - 134
52 - - 109
52 68 - 142
52 69 - 143
52 - - 111
52 - - 130
52 - - 114
52 69 - 140
52 - - 120
52 65 - 177
52 67 - 164
52 - - 115
52 - - 105
52 66 - 155
52 69 - 143
52 65 - -
53 - - 102
53 - - 127
53 69 - 143
53 71 - 139
53 67 - -
53 73 - 133
53 66 - -
53 74 - 135
53 - - 111
53 70 - 142
53 66 - 178
53 66 - 157
53 - - 112
- 53 67 - 161
54 68 102 -
54 - - 98
54 - - 110
54 - - 116
54 - - 131
54 72 - 143
54 - - 132
54 69 - 149
54 - - 99
54 - - 102
54 - - 104
54 - - 118
54 - - 127
54 - - 114
54 - - 104
55 74 - 141
55 - - 108
55 71 - 161
55 70 - -
55 72 - 155
-- 55 - - 121
55 74 - 135
55 - - 119
55 - - 127
55 77 - 140
55 - - 108
55 68 - -
55 - - 101
55 70 - -
55 - - 97
55 - - 102
55 68 - -
55 - - 106
56 74 - 143
56 - - 143
56 69 - -
56 72 - 152
56 74 - 140
56 73 - 161
56 71 - 158
56 - - 107
56 - - 98
56 - - 97
56 - - 121
56 - - 106
56 70 - -
56 73 - 164
56 73 - 142
56 - - 99
56 - - 100
56 - - 109
56 - - 122
56 - - 137
57 75 - 145
57 - - 94
57 - - 133
57 72 - -
57 76 - 143
57 - - 126
57 - - 124
57 - - 112
57 - - 130
57 - - 142
57 - - 132
57 - - 114
57 - - 102
57 72 - -
57 - - 90
- 57 75 - 166
57 - - 108
58 - - 129
58 - - 91
58 78 - 141
58 - - 110
58 77 - 139
58 74 - -
58 - - 93
58 - - 111
58 - - 143
58 - - 88
58 74 - 168
58 - - 142
58 - - 90
58 75 - 146
- 58 74 - -
- 58 74 - -
58 - - 90
58 73 - -
59 - - 123
59 - - 116
59 - - 87
59 - - 97
- 59 78 - 148
59 75 - 167
59 - - 97
59 - - 90
59 - - 111
59 - - 139
59 - - 96
-59 76 - -
59 - - 98
59 - - 114
59 - - 88
60 - - 123
60 - - 100
60 - - 107
60 - - 100
60 - - 112
60 - - 139
60 81 - 151
60 - - 111
60 - - 120
60 - - 125
60 74 - -
60 - - 104
60 - - 118
60 77 - 163
60 - - 127
- 60 - - 85
60 - - 140
60 - - 125
60 - - 97
61 - - 127
61 - - 125
61 - - 82
61 - - 138
61 78 - 164
61 79 - 179
61 - - 103
61 - - 87
61 - - 121
61 - - 134
62 82 - 167
62 - - 143
62 - - 93
62 83 - 173
62 - - 138
62 - - 91
62 79 - 175
62 - - 96
62 85 - 160
62 - - 84
62 - - 96
62 - - 123
62 - - 139
62 - - 122
62 77 - -
62 - - 103
62 - - 116
62 - - 99
62 - - 104
62 - - 142
62 - - 111
62 80 - -
62 - - 135
63 81 - -
63 - - 142
63 85 - 171
63 - - 133
63 - - 148
63 - - 85
63 89 - 155
63 - - 135
63 - - 144
63 - - 86
63 86 - 158
63 - - 108
64 79 - -
64 - - 127
64 85 - -
64 85 - -
64 83 - -
64 - - 132
-- 64 88 - 160
64 81 - 172
64 - - 118
65 - - 129
65 81 - -
65 - - 117
65 91 - 164
65 - - 154
65 - - 147
65 - - 149
65 - - 129
65 - - 117
65 82 - -
65 - - 150
- 65 - - 125
65 - - 116
65 - - 120
65 - - 105
65 - - 78
65 - - 97
65 - - 92
65 - - 93
65 - - 146
65 82 - -
65 - - 146
65 - - 144
66 88 - 176
66 91 - 170
66 - - 76
66 89 - -
66 82 - -
66 - - 110
66 - - 115
66 - - 102
66 - - 101
66 - - 134
66 - - 84
67 - - 136
67 85 - -
67 - - 137
67 - - 76
67 - - 125
67 - - 92
67 - - 163
67 88 - -
67 - - 119
67 - - 144
67 - - 75
67 - - 131
67 - - 125
67 96 - 162
67 - - 116
67 - - 160
68 - - 115
68 - - 167
68 - - 110
68 93 - 172
68 - - 142
68 95 - -
68 - - 135
68 - - 107
68 - - 100
68 100 - 173
68 - - 152
- 68 - - 157
68 88 - 169
68 - - 145
68 - - 139
69 - - 152
69 - - 73
69 97 - 179
69 - - 129
69 - - 97
69 - - 161
69 96 - 164
69 - - 162
69 - - 159
69 - - 174
69 - - 128
69 - - 153
69 - - 127
69 90 - -
69 - - 153
69 95 - -
69 - - 117
70 87 - -
70 - - 159
70 - - 122
70 - - 156
70 - - 89
70 - - 95
70 - - 144
70 - - 156
70 - - 141
70 - - 139
70 - - 100
70 - - 103
70 - - 160
70 - - 105
70 - - 100
70 - - 136
71 92 - -
71 98 - -
71 - - 153
71 - - 171
71 - - 178
71 - - 100
71 - - 101
71 - - 99
71 - - 137
71 93 - -
71 105 - 177
72 100 - -
- 72 93 - -
72 94 - -
72 99 - -
72 - - 144
72 - - 148
72 - - 138
72 - - 148
72 - - 136
72 - - 150
72 - - 156
73 - - 144
73 - - 116
73 - - 131
73 - - 124
() 73 105 - 175
73 - - 136
73 - - 138
73 - - 145
73 - - 115
73 - - 115
- 73 - - -
73 - - 141
73 - - 127
73 - - 126
73 - - 130
- 73 - - 155
73 - - 140
73 96 - -
73 - - 159
73 - - 175
73 - - 161
73 - - 138
74 - - 104
74 - - 148
74 - - 122
- 74 - - 131
74 - - 129
74 - - 155
74 - - 149
74 - - -
74 - - 148
74 - - 123
74 - - 120
75 - - 131
75 - - 143
75 - - 164
75 - - 155
75 - - 111
- 75 - - 174
75 - - 162
75 101 - -
75 - - 113
76 - - 152
76 - - 160
76 - - 138
76 - - 107
76 - - -
76 - - 119
76 - - 151
76 - - 158
76 - - 104
76 100 - -
76 106 - -
76 108 - -
76 - - 149
76 - - -
76 - - 153
77 - - 142
77 117 - -
77 - - 146
77 - - -
77 - - -
77 - - 175
77 - - -
77 - - 173
77 110 - -
77 - - 131
77 - - 172
77 - - 141
77 - - 131
77 - - 174
77 - - 141
77 - - 144
78 - - 115
78 - - 122
78 - - 146
78 - - 168
78 - - 138
78 - - 139
78 - - 123
78 - - -
78 - - 134
78 - - 153
78 - - 131
78 - - 178
78 - - 123
78 - - 136
78 - - 121
79 - - 169
- 79 - - -
79 - - 142
79 - - 128
79 - - 164
79 - - 152
79 - - 132
79 - - 139
79 - - 109
79 - - 132
79 108 - -
79 - - 158
79 - - 119
79 - - 162
80 - - 130
80 - - 161
80 117 - -
80 - - -
80 - - 147
80 - - 167
80 - - 128
80 - - 153
80 - - 148
80 - - 151
80 - - -
80 - - 160
80 - - -
80 - - 139
80 - - -
80 116 - -
81 - - 145
81 - - 126
- 81 - - 138
81 - - -
81 - - 146
81 - - 137
81 108 - -
81 - - 125
- 81 - - 121
81 - - 167
82 - - 132
82 130 - -
82 - - 140
82 - - 176
82 - - 159
82 - - 119
82 - - 159
82 - - 176
82 - - -
82 - - 129
83 - - -
83 - - -
83 - - 170
83 - - -
83 - - 134
83 - - -
83 - - 177
83 - - 140
83 - - 156
83 - - 140
84 - - 143
84 - - 161
84 - - 131
84 - - 116
84 - - 166
84 - - 95
84 132 - -
() 85 - - 174
85 135 - -
-- 85 117 - -
85 - - 127
85 - - 141
85 - - 172
85 - - 122
- 86 - - -
86 - - 134
86 - - 179
86 - - -
86 - - 128
86 - - 142
86 - - 178
86 - - 145
87 117 - -
87 - - -
87 - - 147
87 - - 140
87 - - -
87 - - 99
87 - - 137
87 126 - -
87 - - 176
88 - - -
88 - - 129
- 88 - - 114
88 - - 133
88 - - 144
88 - - 174
88 - - 122
88 131 - -
88 - - 133
88 - - 149
88 - - 148
88 - - 156
88 - - 163
88 - - 144
88 - - 179
89 - - 124
89 - - 137
- 89 - - -
90 - - 179
90 - - -
90 - - 153
90 - - 140
90 148 - -
90 - - 174
90 - - -
- 90 - - 161
91 - - 175
91 - - -
91 - - 170
91 - - 124
91 - - 145
91 - - 170
- 91 - - -
91 - - 154
92 - - -
92 - - -
92 - - 148
92 - - 149
92 - - 178
92 - - 168
92 - - 129
92 - - -
92 - - 140
92 - - 134
93 - - 153
93 - - 104
93 - - -
93 - - -
93 - - 139
94 - - -
94 - - 138
94 143 - -
95 - - 155
95 - - -
95 - - 152
95 - - 145
95 - - 112
95 - - -
95 - - 129
95 - - -
95 - - 129
95 - - -
96 - - -
96 - - 150
96 - - 178
96 - - -
97 - - 168
97 - - -
97 - - 115
97 - - -
98 - - 179
98 - - -
98 - - -
98 - - 137
98 - - 172
98 - - -
99 - - 159
99 - - -
99 - - 174
99 - - 158
100 - - 113
100 - - -
100 - - -
-- 100 - - 175
100 - - 144
101 - - -
102 - - 162
102 - - -
102 - - 132
103 - - 147
103 - - -
103 - - -
103 - - -
() 103 - - -
104 - - -
104 - - -
105 - - 112
105 - - -
105 - - -
105 - - 135
105 - - -
106 - - 147
106 - - 174
107 - - -
107 - - -
107 - - -
109 - - -
109 - - -
109 - - -
110 - - 140
110 - - -
111 - - 160
-- 111 - - -
112 - - -
114 - - 172
115 - - -
117 - - 137
117 - - -
121 - - -
122 - - -
122 - - -
- 123 - - -
- 123 - - -
124 - - 144
125 - - -
- 131 - - -
133 - - -
133 - - -
134 - - -
135 - - 155
136 - - -
- 141 - - -
156 - - -
163 - - -
168 - - -
172 - - -
173 - - -
The model also has Aldan, Argun, Arsk, Artyshta, Artyom, Artyomovsk, Ayan, Babushkin, Bagdarin, Baykalsk, Baykit, Batagay, Belaya Gora, Bely, Birch, Bogorodskoye, Bolotnoye, Buinsk, Vanavara, Vanino, Veliky Ustyug, Verkhnevilyuys, Verkhoturie Vilyuysk Vyazemskij, Gremyachinsk, Gousinoozerskaya, Davlekanovo, Deputatsky, Dixon, Dolinsk, Erbogachen, Zhigansk, Zhukovka Zabaykalsk, Zavitinsk, Zainsk, Zalahtove, Zarinsk, Zlynka, Zuevka, Zyrianka, Igarka Igrim Izberbash, Izborsk Lime, Kazachinskoye, Kalachinsk, Kalevala, Karagaysky, Karasuk, Karachev, Kargasok, Kargat, Keperveem, Kinel, Kirensk, Kola, Koslan, Red Chikoy, Kupino, Kurilsk, Leninsk, Lobnya, Luza, Lyuban, Makarov, Makushino, Malarkhangelsk, Maloarkhangelsk, Maloarkhangelsk , Meshchovsk, Minyar, Mogocha, Murashi, Myski, Mytishchi, Nazyvaevsk, Nartkala, Nizhneangarsk, Novoabzakovo, Novorzhev, Novosil, Nyurba, Obluchye,Ob, Lake Karachinskoye, Ozersk, Olenyok, Olekminsk, Omolon, Omsukchan, Okhotsk, Pavelets, Pevek, Peno, Petukhovo, Plyos, Poronaysk, Przhevalskoye, Yarn, Sakkyryr, Salmi, Saskylakh, Svetlogorsk (Krasnoyarsk, Yekaterinburg). Seymchan, Simeiz, Blue Sedge, Slyudyanka, Sol-Iletsk, Sonkovo, Sorsk, Sosnovka, Spas-Demensk, Srednekolymsk, Sretensk, Suntar, Susuman, Taksimo, Talakan, Terek, Tiksi, Toguchin, Tolka, Tommot, Topruki, Topki, Topki , Turukhansk, Ust-Kachka, Ust-Koksa, Ust-Kuyga, Ust-Nera, Ust-Tsilma, Khandyga, Khatanga, Khilok, Hill, Honuu, Khotynets, Chara, Chemal, Cheremkhovo, Cherepanovo, Chernoluchye, Chersky, Chokurdym, Chokurdy, Chulurdy , Chumikan, Chormoz, Shilka, Evensk, Yuzhno-Kurilsk - these cities either become infected later than the simulated period, or do not gain 1000 infected.Peno, Petukhovo, Ples, Poronaysk, Przhevalskoye, Yarn, Sakkyryr, Salmi, Saskylakh, Svetlogorsk (Krasnoyarsk.), Severo-Yeniseysk, Seymchan, Simeiz, Blue Osoka, Slyudyanka, Sol-Iletsk, Sonskovo, Sorsk-Spensk, Sosnovka , Srednekolymsk, Sretensk, Suntar, Susuman, Taksimo, Talakan, Terek, Tiksi, Toguchin, Only, Tommot, Firebox, Tour, Turan, Turukhansk, Ust-Kachka, Ust-Koks, Ust-Kuyga, Ust-Nera, Ust-Tsilma , Khandyga, Khatanga, Khilok, Hill, Honuu, Khotynets, Chara, Chemal, Cheremkhovo, Cherepanovo, Chernuluchie, Chersky, Chokurdah, Chulym, Chumikan, Cheremoz, Shilka, Evensk, Yuzhno-Kurilsk - these cities either become infected later than the simulated period, or Do not gain 1000 infected.Peno, Petukhovo, Ples, Poronaysk, Przhevalskoye, Yarn, Sakkyryr, Salmi, Saskylakh, Svetlogorsk (Krasnoyarsk.), Severo-Yeniseysk, Seymchan, Simeiz, Blue Osoka, Slyudyanka, Sol-Iletsk, Sonskovo, Sorsk-Spensk, Sosnovka , Srednekolymsk, Sretensk, Suntar, Susuman, Taksimo, Talakan, Terek, Tiksi, Toguchin, Only, Tommot, Firebox, Tour, Turan, Turukhansk, Ust-Kachka, Ust-Koks, Ust-Kuyga, Ust-Nera, Ust-Tsilma , Khandyga, Khatanga, Khilok, Hill, Honuu, Khotynets, Chara, Chemal, Cheremkhovo, Cherepanovo, Chernuluchie, Chersky, Chokurdah, Chulym, Chumikan, Cheremoz, Shilka, Evensk, Yuzhno-Kurilsk - these cities either become infected later than the simulated period, or Do not gain 1000 infected.Sonkovo, Sorsk, Sosnovka, Spas-Demensk, Srednekolymsk, Sretensk, Suntar, Susuman, Taksimo, Talakan, Terek, Tiksi, Toguchin, Tolka, Tommot, Firebox, Tour, Turan, Turukhansk, Ust-Kachka, Ust-Koks, Ust- Kuyga, Ust-Nera, Ust-Tsilma, Khandyga, Khatanga, Khilok, Hill, Honuu, Khotynets, Chara, Chemal, Cheremkhovo, Cherepanovo, Chernoluchye, Chersky, Chokurdah, Chulym, Chumikan, Chermoz, Shilka, Evensk, Yuzhno-Kul these cities either become infected later than the simulated period, or do not gain 1000 infected.Sonkovo, Sorsk, Sosnovka, Spas-Demensk, Srednekolymsk, Sretensk, Suntar, Susuman, Taksimo, Talakan, Terek, Tiksi, Toguchin, Tolka, Tommot, Firebox, Tour, Turan, Turukhansk, Ust-Kachka, Ust-Koks, Ust- Kuyga, Ust-Nera, Ust-Tsilma, Khandyga, Khatanga, Khilok, Hill, Honuu, Khotynets, Chara, Chemal, Cheremkhovo, Cherepanovo, Chernoluchye, Chersky, Chokurdah, Chulym, Chumikan, Chermoz, Shilka, Evensk, Yuzhno-Kul these cities either become infected later than the simulated period, or do not gain 1000 infected.Shilka, Evensk, Yuzhno-Kurilsk - these cities either become infected later than the simulated period, or do not get 1000 infected.Shilka, Evensk, Yuzhno-Kurilsk - these cities either become infected later than the simulated period, or do not get 1000 infected.Scenario 2
The most "isolated" scenario. Only 10% of traffic is left in it, that is, getting from one point to another becomes much harder, but possible. This is not a complete quarantine, but a significant reduction in connectivity. People inside cities do not go to fairs and try to isolate themselves, but they do this not ideally, but to the best of their understanding. A period of 180 days is simulated (accordingly, Barnaul, which scored a thousand patients on the 174th day of the model, will not have time to score less than a thousand patients in the process of “crushing” the epidemic). Cities that are not in the table do not recruit more than 1000 patients or do not become infected. >1000 >10.000 >100.000 <1000
18 33 51 -
40 60 - -
42 - - 158
45 66 - -
46 68 - -
46 67 - -
- 46 62 88 -
-- 56 76 - -
61 91 - -
62 86 - -
63 88 - -
- 65 - - 172
65 91 - -
66 94 - -
66 93 - -
() 68 99 - -
69 - - -
69 108 - -
69 - - 160
70 103 - -
70 103 - -
70 104 - -
71 105 - -
71 104 - -
71 105 - -
72 109 - -
- 72 - - -
72 110 - -
72 109 - -
72 112 - -
73 108 - -
73 - - -
74 113 - -
75 118 - -
75 - - -
77 - - -
77 - - -
77 128 - -
78 - - -
80 - - -
81 - - -
81 - - -
81 - - -
82 - - -
82 - - -
82 - - -
83 166 - -
83 - - 168
84 - - -
84 - - 173
84 - - -
84 - - 149
84 178 - -
85 - - -
85 - - 157
85 - - -
85 - - -
86 - - -
86 - - -
86 - - -
87 - - -
88 - - 175
89 - - -
90 - - 155
- 90 - - -
91 - - -
91 - - -
92 - - 141
92 - - -
92 - - 117
92 - - -
93 - - 160
94 - - 175
94 - - 126
94 - - -
94 - - -
94 - - -
94 - - 147
94 - - -
95 - - 157
96 - - -
96 - - -
96 - - -
97 - - -
97 - - -
97 - - -
98 - - -
98 - - -
98 - - 178
98 - - -
99 - - -
99 - - 167
99 - - -
99 - - -
99 - - -
99 - - -
99 - - -
99 - - -
99 - - 135
99 - - -
100 - - -
100 - - -
100 - - -
100 - - -
101 - - -
101 - - -
102 - - 171
102 - - -
103 - - -
103 - - 161
103 - - -
104 - - 128
104 - - -
104 - - -
104 - - -
104 - - -
104 - - -
105 - - 165
105 - - -
105 - - -
105 - - -
- 105 - - 154
107 - - -
107 - - -
107 - - -
108 - - -
108 - - -
109 - - -
110 - - -
111 - - -
111 - - -
111 - - -
112 - - -
113 - - -
113 - - 169
114 - - -
114 - - -
116 - - -
117 - - -
118 - - -
118 - - -
118 - - -
119 - - -
119 - - -
120 - - 174
122 - - -
124 - - -
125 - - -
126 - - -
126 - - -
127 - - -
127 - - -
128 - - -
128 - - -
128 - - -
129 - - -
130 - - -
131 - - -
131 - - -
133 - - -
133 - - -
135 - - -
135 - - -
- 137 - - -
141 - - -
141 - - -
145 - - -
148 - - -
150 - - -
151 - - -
157 - - -
161 - - 176
164 - - -
164 - - -
173 - - -
174 - - -
I remind you once again that this is a simulation of the spread of infection between cities, and not a consideration of mortality and the rest. You can help ODS by writing in a personalmephistopheies.More important: the Skoltech center decided to allocate computing time on the Zhores supercomputer to simulate tasks that are somehow related to containing the epidemic. They have CPU-nodes (a small part, but there are) and GPU-nodes (in the nodes there are 4 Tesla P100). If you have a code parallelized for MPI CPU or need to calculate models in parallel on a GPU, this is suitable. You can bring any tasks that are somehow related to the fight against the epidemic, for example, molecular modeling, traffic flow modeling, etc. Contacts: sergey.sosnin@skoltech.ru or you can apply directly .What does it mean?
Russian logistics is one of the most complex in the world. We have a lot of isolated cities, the population is significantly scattered around the planet, huge empty territories. From tightly connected in the west, the count turns into loosely coupled in the east.This may mean that it is possible to cut off significant parts of the graph from the main foci (but it seems to be too late due to returning to different large cities from abroad).Models without taking into account such a distribution are very much limited by the lack of necessary data.Now the factor of population unevenness means that you can dramatically change the course of the epidemic, blocking transport links.The main commercial conclusion is that the lockdown is now cheaper than the consequences further.I invite you not to take my word for it, but to crack up a model, dig deeper into CSV and code, twist and improve the result that ODS provided. Actually, the community is currently working on refinements to the model.And I emphasize that the fact that the community took on the project, made it and opens the result is a huge precedent in favor of the fact that the data themselves need to be opened. Because sometimes they are extremely useful, even if you do not have the time or the ability to do something with them.Links in one place , a large post with a bunch of links to different studies,mephistopheies- your entry into ODS if you want to help. ODS post bygrismewith model internals and assumptions, a link to the repository .