Li, Li, Chen and Sun TRB 2016 Annual Meeting Bridge Deterioration Prediction Using Markov-Chain Model Based 1 on the Actual Repair Status in Shanghai 2 3 4 Li LI 5 Associate Professor 6 Key Laboratory of Road and Traffic Engineering of the Ministry of Education 7 Tongji University 8 No. 4800 Cao’an Road, Jiading District, 201804, Shanghai, China 9 TEL:(+8621) 69580491, E-mail: [email protected]10 11 Feng LI 12 Associate Professor 13 Key Laboratory of Road Structure and Material of the Ministry of Transport 14 Research Institute of Highway, Ministry of Transport 15 No. 8 Xitucheng Road, Haidian District, 100088, Beijing, China 16 TEL:(+8610) 62079161, E-mail: [email protected]17 (Corresponding Author) 18 19 Zhang CHEN 20 Associate Professor 21 Key Laboratory of Road and Traffic Engineering of the Ministry of Education 22 Tongji University 23 No. 4800 Cao’an Road, Jiading District, 201804, Shanghai, China 24 TEL:(+8621) 69580491, E-mail: [email protected]25 26 Lijun SUN 27 Professor 28 Key Laboratory of Road and Traffic Engineering of the Ministry of Education 29 Tongji University 30 No. 4800 Cao’an Road, Jiading District, 201804, Shanghai, China 31 TEL:(+8621) 69583810, E-mail: [email protected]32 33 34 35 Submitted for Presentation and Publication at the 2016 Annual Meeting of the 36 Transportation Research Board 37 38 39 Submission date: July 31, 2015 40 Revision date: November 15, 2015 41 Word count: 6977 with 3 tables and 4 figures 42
17
Embed
Bridge Deterioration Prediction Using Markov-Chain …docs.trb.org/prp/16-4744.pdf · 1 Bridge Deterioration Prediction Using Markov-Chain Model ... 54 different characteristics of
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
Bridge Deterioration Prediction Using Markov-Chain Model Based 1
on the Actual Repair Status in Shanghai 2
3
4
Li LI 5
Associate Professor 6
Key Laboratory of Road and Traffic Engineering of the Ministry of Education 7
Tongji University 8
No. 4800 Cao’an Road, Jiading District, 201804, Shanghai, China 9
The prediction results of all three bridge components and the whole bridge for the two 347
conservatism regimes are displayed in Figure 2 and Figure 3. The prediction period is 10 years 348
(i.e., 2014–2024). 349
350
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
351
Figure 2 Bridge prediction results under routine maintenance and minor repair. Letters A, B, 352
C, D and E represent bridge grades according to the Chinese technical regulation. 353
354
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
355
Figure 3 Bridge prediction results under medium and major repair. Letters A, B, C, D and E 356
represent bridge grades according to the Chinese technical regulation. 357
358
CR prediction for a new bridge 359
The deterioration process under different conservation regimes can be predicted using the 360
Markov-chain model above for a new bridge with intact condition status in Shanghai. The 361
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
calculation method is described in Equation (9). In Equation (9), 𝐶(𝑡)∗ is the bridge CR grade 362
(status value) after t years; 𝐶(0)∗ is the status vector of the initial condition; 𝑃𝑚𝑎𝑖𝑛𝑡𝑎𝑖𝑛 is the 363
TPM in certain conservation circumstance; and CR is the bridge condition rating vector, which is a 364
constant vector in the case of new bridge. In this study CR= [5; 4; 3; 2; 1]. 365
366
𝐶(𝑡)∗ = 𝐶(0)∗ × (𝑃𝑚𝑎𝑖𝑛𝑡𝑎𝑖𝑛 )𝑡 × 𝐶𝑅 (9) 367
368
It was assumed that new bridges will deteriorate from the intact state during the forecast 369
period of 20 years. Thus, the initial status vector 𝐶(0)∗ was [1, 0, 0, 0, 0]. The bridge condition 370
deterioration tendency for different components and for the whole bridge can be found in Figure 4, 371
in which the ordinate scale 5 to 1 represents CR grades A to E, respectively. 372
373
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
374
Figure 4 Deterioration prediction over 20 years for a new bridge in Shanghai. The ordinate 375 scale 5 to 1 (y axis) represents CR grades A to E, respectively. 376
377
DISCUSSION 378
Figure 2 and Figure 3 show that the bridge condition will deteriorate rapidly if only routine 379
maintenance and minor repair are implemented. Comparatively, the bridge deck system has the 380
fastest decline of the three bridge components analyzed. At the end of the prediction period (2024) 381
the proportion of bridge deck systems with grade A will be less than 20%, while the proportion of 382
grade E deck systems will be as much as 20%. This finding is quite worrying because the safety 383
rating of a grade E bridge is “dangerous”. Medium and major repairs seem to have a significant 384
positive impact on the bridge deck status in the central city, as shown in Figure 3(a) to Figure 3(f), 385
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
but this positive impact is not apparent for the suburban bridges. The latest data in the Shanghai 386
BMS show that the proportion of suburban bridges is 55.6% (2014), thus their conditions have a 387
large impact on the overall state of transport infrastructure in the city. Because the medium and 388
major repair rate is too low in suburban areas of Shanghai (see Table 3), the bridge conditions in 389
these areas cannot be improved significantly. The cases for the superstructure, substructure and the 390
whole bridge are almost the same as for the bridge deck system. Bridges in the central city always 391
have a better performance than their suburban counterparts, and the suburban bridges reduce the 392
overall bridge quality level of the entire city, reflecting that investment on the repair of suburban 393
bridges in Shanghai is very insufficient. Even in the central city, the present repair rate cannot 394
fundamentally improve the technical condition of existing bridges. 395
The results portrayed in Figure 2 and Figure 3 indicate that bridges in Shanghai have entered 396
a rapid deterioration stage, and if the repair efforts cannot be strengthened from now, the future 397
state of these bridges will be quite worrying. This conclusion is also consistent with the empirical 398
findings of the Shanghai bridge management investigation conducted as part of this study. 399
Figure 4 displays the predicted deterioration of a new bridge based on the present bridge 400
conservation strategy in Shanghai. The medium and major repair efforts at present seem to have a 401
significant impact on the bridge deck system and the superstructure, as shown in Figure 4(a) and 402
Figure 4(b), and the impact will be apparent in 6–8 years. Regarding routine maintenance and 403
minor repair, the location of a new bridge seems to have no influence on the bridge deterioration 404
process. However, in the case of medium and major repair, the location effect is obvious and the 405
performance decline of new suburban bridges is predicted to be far more rapid than that of new 406
bridges in the central city. As with existing bridges, the present repair efforts are insufficient to 407
improve the state of the substructures and the whole bridges, as shown in Figure 4(c) and Figure 408
4(d). 409
410
CONCLUSIONS 411
Bridge condition data spanning approximately 11 years in the Shanghai BMS were used in this 412
study to calibrate a Markov-chain model to forecast the bridge condition, and also to find the 413
deterioration tendency, of urban bridges in Shanghai. The bridge CR grades were used to generate 414
the state vector space and the time step was set to one year, matching the bridge inspection 415
frequency and simplifying the form of Markov TPMs. 416
The modeling process considered two conservation strategies, and the actual repair rate was 417
obtained through a conservation survey. The influence of bridge location also was considered. 418
Results from the study support the following conclusions. 419
Bridge conservation efforts (or investment levels) are uneven across the city, with those in the 420
central city being much better than those in suburban areas, although the proportion of suburban 421
bridges is higher (55.6%). Furthermore, the present level of bridge conservation efforts in central 422
Shanghai is insufficient, and the condition in suburban areas is even worse. The medium and 423
major repair efforts at present have a significant positive impact on the deck system and the 424
superstructure of central city bridges, but the impact of these efforts on the substructure and the 425
whole bridge is quite small, indicating that the present conservation strategy cannot improve the 426
bridge conditions fundamentally. The positive impacts of conservation efforts on central city 427
Li, Li, Chen and Sun
TRB 2016 Annual Meeting
bridges do not extend to suburban bridges. Overall, bridges in Shanghai have entered a rapid 428
deterioration stage, and existing repair efforts need to be strengthened immediately. 429
430
ACKNOWLEDGEMENTS 431
This study was funded by The Ministry of Science and Technology of China (Grant Number: 432
2013DFA81910) and National Natural Science Foundation of China (Grant Number: 51378242 433
and 51508411). The authors would like to acknowledge this financial support. 434
435
REFERENCES 436
[1] Technical code of maintenance for city bridges, China, CJJ 99-2003 437
[2] Morcous G. and Hatami, A. Developing deterioration models for Nebraska bridges. Project 438
No. SPR-P1(11) M302, Final Report, Nebraska Department of Roads, Lincoln, NE, 2011 439
[3] Chen Z. Research on technology structure of transportation infrastructure management system. 440