Estimation of financial needs and condition of infrastructure assets
2024-04-15
Example: Reinforced concrete bridges, in terms of surface area per state.
CS1 | CS2 | CS3 | CS4 | CS5 | Total |
---|---|---|---|---|---|
6’878 | 4’149 | 3’186 | 2’570 | 0 | 16’782 |
Once you know the objects you are considering, you can also determine their deterioration transition matrix
\(CS^{t}\) | \(CS^{t+1}\) | ||||
1 | 2 | 3 | 4 | 5 | |
1 | 0.956 | 0.044 | 0 | 0 | 0 |
2 | 0 | 0.968 | 0.032 | 0 | 0 |
3 | 0 | 0 | 0.956 | 0.044 | 0 |
4 | 0 | 0 | 0 | 0.944 | 0.056 |
5 | 0 | 0 | 0 | 0 | 1 |
Consider the maintenance intervention strategies as follows:
IS | CS1 | CS2 | CS3 | CS4 | CS5 |
---|---|---|---|---|---|
1 | Do nothing | Do nothing | Do nothing | Rehabilitation | Replacement |
2 | Do nothing | Do nothing | Do nothing | Do nothing | Replacement |
In connection to the effectiveness vectors and unit costs of interventions
CS | Interventions | CS |
Maintenance unit costs (\(mu/m^2\)) |
Travel time unit costs (\(mu/m^2\)) |
Accident unit costs (\(mu/m^2\)) |
||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | |||||
1 | Do nothing | 0.96 | 0.04 | 0 | 0 | 0 | 0 | 0 | 0 |
2 | Do nothing | 0 | 0.97 | 0.03 | 0 | 0 | 10 | 0 | 0 |
3 | Do nothing | 0 | 0 | 0.96 | 0.04 | 0 | 50 | 0 | 0 |
4 | Do nothing | 0 | 0 | 0 | 0.94 | 0.06 | 100 | 1000 | 100 |
Rehabilitation | 0.75 | 0.25 | 0 | 0 | 0 | 1000 | 2000 | 200 | |
Replacement | 1 | 0 | 0 | 0 | 0 | 3500 | 5000 | 500 | |
5 | Replacement | 1 | 0 | 0 | 0 | 0 | 3500 | 5000 | 500 |
Translation into the deterioration-intervention matrices
CS | 1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|---|
1 | 0.96 | 0.4 | 0 | 0 | 0 |
2 | 0 | 0.97 | 0.03 | 0 | 0 |
3 | 0 | 0 | 0.96 | 0.04 | 0 |
4 | 0.75 | 0.25 | 0 | 0 | 0 |
5 | 1 | 0 | 0 | 0 | 0 |
\[P(S^{t+1}=j \mid S^t=i, S^{t-1}=i^{t-1},\ldots, S^1=i^1, S^0=i^0) \] \[= P(S^{t+1}=j \mid S^t=i) \] \[= P_{ij}\]
A one-step transition matrix, \(P\), with the same dimention as the deterioration matrix is needed, i.e., \(P_{K\times K}\)
\(S^t\) | \(S^{t+1}\) | ||||
1 | 2 | 3 | \(\cdots\) | k | |
1 | \(p_{11}\) | \(p_{12}\) | \(p_{13}\) | \(\cdots\) | \(p_{1k}\) |
2 | \(p_{21}\) | \(p_{22}\) | \(p_{23}\) | \(\cdots\) | \(p_{2k}\) |
3 | \(p_{31}\) | \(p_{32}\) | \(p_{33}\) | \(\cdots\) | \(p_{3k}\) |
\(\vdots\) | \(\vdots\) | \(\vdots\) | \(\vdots\) | \(\vdots\) | \(\vdots\) |
k | \(p_{k1}\) | \(p_{k2}\) | \(p_{k3}\) | \(\cdots\) | \(p_{kk}\) |
\(\quad p_{ij} \geq 0 \qquad \qquad\) \(\sum_{j=1}^{K} p_{ij}=1\)
Steady State Probabilites: \[ \lim_{n \rightarrow \infty} p_{ij}^n = \pi_j \quad ; \quad \pi_j \geq 0 \quad\textbf{&} \quad \sum_{k=1}^{K} \pi_j =1 \]
Steady State Probabilites: \[ \lim_{n \rightarrow \infty} p_{ij}^n = \pi_j \quad ; \quad \pi_j \geq 0 \quad\textbf{&} \quad \sum_{k=1}^{K} \pi_j =1 \]
Steady states are derermined using the follownig formula: \[ \pi_j = \sum_{k=1}^{K} \pi_i \cdot p_{ij} \qquad \forall\quad j=1,2,3,\cdots,k \quad\backepsilon \sum_{k=1}^{K} \pi_i=1 \]
CS | Contributing bits to \(\pi_j\) | ||||
1 | 2 | 3 | 4 | 5 | |
1 | 0.226 | 0.010 | 0 | 0 | 0 |
2 | 0 | 0.420 | 0.014 | 0 | 0 |
3 | 0 | 0 | 0.302 | 0.014 | 0 |
4 | 0.010 | 0.003 | 0 | 0 | 0 |
5 | 0 | 0 | 0 | 0 | 0 |
CS | Contributing bits to \(\pi_j\) | \(\pi_j\) | ||||
1 | 2 | 3 | 4 | 5 | ||
1 | 0.226 | 0.010 | 0 | 0 | 0 | 0.237 |
2 | 0 | 0.420 | 0.014 | 0 | 0 | 0.434 |
3 | 0 | 0 | 0.302 | 0.014 | 0 | 0.316 |
4 | 0.010 | 0.003 | 0 | 0 | 0 | 0.014 |
5 | 0 | 0 | 0 | 0 | 0 | 0 |
Sum | 0.237 | 0.434 | 0.316 | 0.014 | 0 | 1 |
If \(N\) equals a large number of transitions, and \(X_i\) is the number of times an asset enters state \(i\)
\[ N \rightarrow\infty\quad,\frac{X_i}{N}\rightarrow \pi_i \] \[ \eta_\infty=\sum_{i=1}^{K}\pi_i\cdot C(i) \qquad\qquad C(i)=C(i)^{bI}+C(i)^{dI} \]
CS | Contributing bits to \(\pi_j\) | \(\pi_j\) | Maintenance Unit Costs \((mu/m^2)\) | Travel Time Unit Costs \((mu/m^2)\) | Accident Unit Costs \((mu/m^2)\) | Total Costs \((mu/m^2)\) | ||||
1 | 2 | 3 | 4 | 5 | ||||||
1 | 0.226 | 0.010 | 0 | 0 | 0 | 0.237 | 0 | 0 | 0 | 0 |
2 | 0 | 0.420 | 0.014 | 0 | 0 | 0.434 | 4.34 | 0 | 0 | 4.34 |
3 | 0 | 0 | 0.302 | 0.014 | 0 | 0.316 | 15.78 | 0 | 0 | 15.78 |
4 | 0.010 | 0.003 | 0 | 0 | 0 | 0.014 | 13.88 | 27.77 | 2.78 | 44.43 |
5 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Sum | 0.237 | 0.434 | 0.316 | 0.014 | 0 | 1 | 1 | 27.77 | 2.78 | 64.55 |
Asset state |
p |
---|---|
1 | 0.237 |
2 | 0.434 |
3 | 0.316 |
4 | 0.014 |
5 | 0.000 |
Sum | 1.000 |
Steady State
Asset state |
Maintenance Unit Costs \((mu/m^2)\) |
Travel Time Costs \((mu/m^2)\) |
---|---|---|
1 | 0.00 | 0.00 |
2 | 4.34 | 0.00 |
3 | 15.75 | 0.00 |
4 | 13.88 | 27.77 |
5 | 0.00 | 0.00 |
Sum | 34.00 | 27.77 |
Asset state |
Accident Costs \((mu/m^2)\) |
Total Costs \((mu/m^2)\) |
---|---|---|
1 | 0.00 | 0.00 |
2 | 0.00 | 4.34 |
3 | 0.00 | 15.78 |
4 | 2.78 | 44.43 |
5 | 0.00 | 0.00 |
Sum | 2.78 | 64.55 |
At least two interventions strategies for each asset subcategory.
Costs \((mu/m^2)\) | IS1 | IS2 |
---|---|---|
Maintenance | 18’782 | 39’030 |
Travel time | 18’510 | 131’263 |
Accident | 1’851 | 13’126 |
Total | 40’142 | 183’419 |
Steady state | IS1 (prob.) | IS1 \((m^2)\) | IS2 (prob.) | IS2 \((m^2)\) |
---|---|---|---|---|
1 | 0.237 | 3’972 | 0.238 | 3’991 |
2 | 0.434 | 7’281 | 0.327 | 5’488 |
3 | 0.316 | 5’296 | 0.238 | 3’991 |
4 | 0.014 | 233 | 0.187 | 3’136 |
5 | 0.000 | 0 | 0.010 | 176 |
Questions?
Yushu An
Doctoral student
yushan@ethz.ch
ETH Zürich
Institute of Construction and Infrastructure Management (IBI)
Chair of Infrastructure Management
HIL G 32.1, Stefano-Franscini-Platz 5
8093 Zürich