Optimization of Life-Cycle Cost and Reliability Allocation for Rail Systems

Transcription

1 Optimization of Life-Cycle Cost and Reliability Allocation for Rail Systems Yung-Cheng (Rex) Lai and Chia-Tsung Lu Department of Civil Engineering National Taiwan University

2 System design considers trade-off between cost and reliability A rail system consists of a number of subsystems, and each has its own cost and reliability Planners have to carefully allocate the reliability and budget by examining the trade-off between cost and reliability Rail System Reliability : 0.9 Cost : 100 billion Infrastructure Reliability : 0.98 Cost : 50 billion Rolling Stock Reliability : 0.97 Cost : 25 billion Signal & Communication Reliability : 0.96 Cost : 15 billion Electricity Reliability : 0.99 Cost : 10 billion 2

3 MTBF does not consider its impact to the customers Various reliability allocation methods have been developed in the past. Weighting Method Optimization Method For rail system, reliability (system reliability) can be defined as the mean time between failures (MTBF) Time between failure (TBF) TBF TBF Exposure Failure Failure Failure Failure However, this attribute does not consider its effect on passengers ( the consequence of failure) 3

4 System Reliability vs. Service Reliability Train failure MTBF: 100,000 train-hour Communication failure MTBF: 100,000 train-hour Failure frequency MTBF System Reliability Consequence of failure Delay Service Reliability 4

5 From System Reliability to Service Reliability Service reliability (e.g. delay or on-time percentage) is more favorable than system reliability because it considers customers satisfaction Service Reliability Planners should balance the trade-off among service reliability, system reliability, and LCC in Rail System Design Rail System Design Life Cycle Cost System Reliability 5

6 Key Elements in Rail System Design LCC LCC for railway systems typically includes capital investment, operating cost, and maintenance cost within the planning period Service Reliability Rail System Design Life Cycle Cost System Reliability LCC = Capital Cost + Operating Cost + Maintenance Cost Employing LCC is more appropriate in decision making than solely employing capital investment Some products have low capital investment but high operating and maintenance costs Others have high capital investment but low operating and maintenance costs 6

7 Key Elements in Rail System Design System Reliability System reliability is defined as MTBF or MDBF Service Reliability Relationship between system reliability and LCC can be obtained from suppliers Life Cycle Cost Rail System Design System Reliability Time or distance between failures Exposure Failure Failure Failure Failure 7

8 Key Elements in Rail System Design Service Reliability Service reliability identifies the effect on passengers Target Service Reliability On-time arrival percentage (with no buffer): proportion of on-time operations in terms of total system operating time (in train-hour) r sys P P ii r sys = On-time arrival percentage P = d i Life Cycle Cost Delay train-hour 100% On-time arrival train-hour Total system operational time (in train-hour) in a defined period d i = Delay (in train-hour) of subsystem i Service Reliability Rail System Design System Reliability 8

9 Optimization Framework Design Service Reliability Investment Alternatives Output Data Minimize Cost Model System Reliability Alternative Evaluator (AE) Investment Alternatives Table Investment Selector (IS) Life Cycle Cost Maximize Reliability Model Service Reliability Demand Design System LCC 9

10 Alternative Evaluator (AE) AE evaluates all possible alternatives and generates an investment alternatives table with their LCC, system reliability, and service reliability Service reliability needs to be computed based on system reliability and operational data Investment Alternatives LCC and System Reliability can be obtained from suppliers Alternative Evaluator (AE) Investment Alternatives Table Demand Operational data is provided by operators 10

11 Alternative Evaluator (AE) AE evaluates all possible alternatives and generates an investment alternatives table with their LCC, system reliability, and service reliability Service reliability needs to be computed based on system reliability and operational data Number of failures T i Dik NQik i I, k K Mik Average train delay D ik = Delay (in train-hour) of subsystem i with alternative k M ik = MTBF or MDBF of subsystem i with alternative k T i = Operational time or distance of subsystem i in a defined period N = Average number of online trains Q ik = Average delay (in hours) from a failure of subsystem i with alternative k 11

12 Alternative Evaluator (AE) AE evaluates all possible alternatives and generates an investment alternatives table with their LCC, system reliability, and service reliability Service reliability needs to be computed based on system reliability and operational data T i Dik NQik i I, k K Mik r sys P ii P d i 100% 12

13 Alternative Evaluator (AE) Investment Alternatives Alternative Evaluator (AE) Demand Investment Alternatives Table Subsystem Alternatives MDBF (M ik ) LCC (C ik ) Delay (D ik ) (i ) (k ) (train-km) billion dollars (train-hours) 1 29, , Train 3 62, , , , ,629 Electricity 3 48, , , , , Track 3 64, , , , ,047 Signal 3 39, , , , ,313 Communication 3 78, , , , Station 3 123,

14 Investment Selector (IS) IS identifies the best alternative for every subsystem according to acceptable LCC or service reliability Minimize Cost Model (MCM): Minimizing total LCC according to acceptable service reliability Maximize Reliability Model (MRM): Maximizing service reliability according to available LCC Design Service Reliability Minimize Cost Model Investment Selector (IS) Maximize Reliability Model Design System LCC Output Data System Reliability Life Cycle Cost Service Reliability 14

15 Minimize Cost Model (MCM) Min ii kk C ik ik Minimize total LCC S.t. kk ik 1 i I d D i I i ik ik kk P d i ii 100% R P Only one alternative can be chosen for a subsystem Compute delay for each subsystem Fulfill the service reliability requirement and ik 0,1 i I, k K d 0 i I i 15

16 Maximize Reliability Model (MRM) Max P ii P d i 100% Maximize service reliability S.t. kk ik d D i I i ik ik kk ii kk 1 i I C ik ik B Only one alternative can be chosen for a subsystem Compute delay for each subsystem Constraint on LCC and ik 0,1 i I, k K d 0 i I i 16

17 Two case studies to demonstrate the potential use Two case studies with empirical data obtained from a rail system in Taiwan were performed to show the potential use of the proposed method Case I : New System Design Designing a new passenger rail system Selecting appropriate alternatives for subsystems according to design service reliability MCM Case II : Existing System Improvements Improving the reliability of an existing rail system Subject to constraint on available increment in LCC MRM 17

18 Case I : New System Design 25-km passenger rail system Estimated demand is 140,000 passengers per day Six subsystems : train, signal, communication, electricity, station, and track Design service reliability (on-time arrival percentage) is from 95% to 99%, with 1% increments Investment Alternatives Subsystem Alternatives (i ) (k ) Train Electricity Track Signal Communication Station MDBF (M ik ) LCC (C ik ) Delay (D ik ) (train-km) (billion dollars) (train-hours) 1 29, , , , , , , , , , , , , , , , , , , , , , , , , , ,

19 High design service reliability results in high MDBF and LCC MCM efficiently solved this problem by using CPLEX within seconds MDBF (train-km) LCC (billion dollars) Subsystem Design Service Reliability 95% 96% 97% 98% 99% Train 29,274 39,799 62, , ,593 Electricity 29,274 29,274 38,521 66, ,907 Track 36,716 56,073 89, , ,915 Signal 29,274 29,274 29,274 31,039 62,192 Communication 36,716 53,899 78, , ,146 Station 84, , , , ,958 Train Electricity Track Signal Communication Station Total

20 MDBF (train-km) Resulting MDBF from MCM The difference in MDBF among subsystems becomes obvious as service reliability level increase 800,000 Train Electricity Track Signal Communication Station 700, , , , , , ,000 Little difference 0 95% 96% 97% 98% 99% Design Service Reliability Large difference 20

21 LCC (billion dollars) Resulting LCC from MCM Increase in total LCC from 95% to 97% is modest but very sharp from 97% to 99% because of the nonlinear relationship between cost and reliability 200 Train Electricity Track Signal Communication Station Total Increase moderately Increase sharply % 96% 97% 98% 99% Design Service Reliability 21

22 Case II : Existing System Improvement Demand is the same as Case I and service reliability of the existing system is 97% with improvement LCC from 1 ~ 5 billions Not all of the subsystems can be easily changed so we consider alternatives for communication, electricity, and track in this case MRM efficiently solved this problem by using CPLEX within seconds Subsystem Alternatives MDBF (M ik ) LCC (C ik ) Delay (D ik ) (i ) (k ) 1 (train-km) 29,274 (billion dollars) (train-hours) 2, , ,629 Electricity 3 48, , , , , Track 3 64, , , , ,313 Communication 3 78, ,589 22

23 In crement of MDBF (train-km) Resulting MDBF from MRM Track has the most significant increase in MDBF, followed by communication and electricity Large difference Electricity Track Communication 30,000 25,000 20,000 15,000 10,000 Little difference 5, Design LCC (billion dollars) 23

24 Increment of LCC (billion dollars) Resulting LCC from MRM More LCC have been allocated to electricity, and communication for all scenarios Electricity Track Communication Impact on delay from communication and electricity failures is more severe than that from track failures Design LCC (billion dollars) 24

25 Conclusions This research develops an optimization framework to assist decision makers in optimally allocating service reliability, system reliability, and LCC It is essential to incorporate service reliability in rail system design The accuracy of the reliability data obtained from suppliers is a key to successfully determine the optimal investment plan through this process Test process and data Actual performance data Certificate of conformity The proposed tool can help operators maximize their return on investment and provide reliable service to their customers 25

26 Thank You! Railway Technology Research Center at National Taiwan University NTU Civil Research Building, Rm 907 No.188, Sec. 3, Xinhai Rd., Da-an Dist., Taipei City 106, Taiwan Telephone Number: Website: 26