Medellín - Minas - Doctorado en Ingeniería - Ingeniería Civil · 2023
Freight-Transit tour synthesis
This research introduces a multi-class demand synthesis model for transit and freight, utilizing entropy maximization and fuzzy logic. The model incorporates traffic data and fuzzy parameters to accommodate uncertainty. The use of fuzzy logic enhances classical modeling by providing flexibility and addressing data uncertainty, a critical aspect in resource-constrained decision-making scenarios. Finite resources such as road capacity necessitate optimal decision-making. Flexible models are essential, as not all constraints can be fully met. Fuzzy logic excels in handling variability and uncertainty, improving results' reliability. It aids in estimating congestion patterns, emissions levels, and accidents, thereby providing valuable insights to decision-makers. Fuzzy logic's flexibility is crucial for real-world adaptability. It enhances transportation planning, benefiting urban mobility. Results' accuracy directly impacts decisions, and fuzzy logic incorporates real-world variability into models. The research focuses on triangular membership functions, a commonly used approach. Fuzzy logic's adaptability is compared with deterministic models, demonstrating superior performance. It helps in finding satisfactory solutions when full constraint satisfaction is unfeasible. Pareto frontiers indicate multi-objective optimization. Decision-makers can use this frontier to choose the right model based on accomplishment versus entropy trade-offs. Fuzzy logic accommodates partial solutions when strict constraints cannot be met. Trials with a developed model show that capacity and cost significantly influence outcomes. Sensitivity analyses reveal the model's robustness. The model's application is promising for shared lanes and infrastructure optimization, handling data variability and uncertainty. It aids in decision-making for urban transportation planning and infrastructure development. Government agencies must strategize mobility elements. Accurate data are crucial for decisions related to routes, traffic management, and infrastructure. Fuzzy logic can guide decisions about shared lanes and resource allocation, enhancing urban transportation planning and development.
Full text 144 pages with text out of 148
Read the full thesis Record in the repository
Contents
- Introductionp. 17
- The aim and objectives of the researchp. 23
- Structure of the thesisp. 23
- Referencesp. 25
- Introductionp. 17
- Backgroundp. 32
- 2.2.1. Freight demand synthesisp. 32
- 2.2.2. Freight OD synthesisp. 34
- 2.2.3. Freight tour synthesis - FTSp. 35
- 2.2.4. Fuzzy Logic (FL) and transportation modelingp. 35
- Modeling formulationp. 37
- 2.3.1. Entropy-based FTS deterministic formulationp. 37
- 2.3.2. Entropy-based FTS formulation with fuzzy parametersp. 39
- Numerical Experimentp. 46
- 2.4.1. Case studyp. 46
- 2.4.2. Modeling approachp. 47
- 2.4.3. Results and analysisp. 48
- Concluding Remarks and Planning Implicationsp. 60
- Referencesp. 25
- Entropy-based Transit Tour Synthesis using Fuzzy Logicp. 69
- Introductionp. 17
- Backgroundp. 32
- 3.2.1. Transit demand synthesisp. 72
- 3.2.2. EM in transportation modelingp. 73
- 3.2.3. Fuzzy logic and membership functions in transportation modelingp. 75
- Estimation of transit tours based on entropy maximization and fuzzy logicp. 77
- 3.3.1. Transit tour synthesis formulation (TTS) using EMp. 78
- 3.3.2. TTS formulation using the fuzzy logic formulationp. 83
- Numerical experiments: TTS using EM-FLp. 87
- Conclusions and discussionp. 100
- Referencesp. 25
- Buses and Trucksp. 107
- Introductionp. 17
- Literature reviewp. 109
- 4.2.1. Freight Tours Transportationp. 109
- 4.2.2. Transit tours transportationp. 110
- 4.2.3. Exclusive or dedicated lanes (Bus and trucks)p. 110
- 4.2.4. Integration between freight and transitp. 112
- 4.2.5. Entropy maximization (EM) in tour transportation modelingp. 113
- FTS - Deterministic and Flexible Modelingp. 113
- TTS - Deterministic and Flexible Modelingp. 114
- Freight transit tour synthesis (FTTS) modelling: Deterministic and Flexible modelsp. 114
- 4.3.1. Link performance function estimationp. 115
- 4.3.2. FTTS Entropy Maximization Formulationp. 117
- 4.3.3. FTTS Entropy Maximization Formulation using fuzzy parametersp. 122
- Numerical experiments of FTTS (GAMS) using FLp. 125
- 4.4.1. Networkp. 29
- 4.4.2. Integrated link performance functionp. 127
- 4.4.3. Analysis and Resultsp. 128
- Scenario 1: Analysis without capacity and cost constraintsp. 128
- Scenario 2: Analysis with capacity constraintp. 133
- Scenario 3: Analysis with cost constraintp. 134
- Concluding Remarksp. 135
- Referencesp. 25
- Conclusions and recommendationsp. 141
- Conclusionsp. 141
- Recommendationsp. 145
- Figure 2-1 Membership function example (triangular)p. 41
- Figure 2-2 Sioux Falls network (Gonzalez-Calderon & Holguín-Veras, 2019)p. 47
- Figure 2-3. Pareto frontier. Entropy vs. accomplishment levelp. 50
- Figure 2-4. Traffic change for each solutionp. 51
- Figure 2-5. Truck tour volumes for different ε valuesp. 52
- Figure 2-6. Accomplishment of the node-productionp. 53
- Figure 2-7. Percentage of the accomplishment node-production constraintp. 54
- Figure 2-8. Percentage of the accomplishment link-volume constraintp. 55
- Figure 2-9. Histogram of % of accomplishment in comparison FL and No FL casesp. 59
- Figure 3-1. Sioux Falls Network. Source: (Wei et al., n.d.)p. 88
- Figure 3-2. Transit Pareto frontier (Entropy vs. Accomplishment level)p. 89
- Figure 3-3. Bus traffic changes for each different solutionp. 90
- Figure 3-4. Bus tour volumes for different ε valuesp. 91
- Figure 3-5. TTS - Level (%) of accomplishment in nodes productionp. 92
- Figure 3-6. TTS - Percentage accomplishment in node-production constraintp. 93
- Figure 3-7. TTS - Percentage accomplishment in link-volume constraintp. 93
- Figure 4-1. Central transitway diagram (BRT)p. 111
- Figure 4-2. Curb side bus lane diagramp. 112
- Figure 4-3. Sioux Falls Networkp. 126
- Figure 4-4. Pareto frontier Entropy vs. Epsilonp. 129
- Figure 4-6. Volume by link - buses and trucks for different solutionsp. 130
- Figure 4-7. Volume by link - buses and trucks for different solutionsp. 131
- Figure 4-8. Volume by tour - buses and trucks for different solutionsp. 132
- Figure 4-9. Variation of the Cost for every epsilon valuep. 133
- Table 2-1. Results of testing statistically differences between all feasible solutionsp. 49
- Table 2-2. Descriptive analysis of the accomplishment node-production constraintp. 54
- Table 2-3. Descriptive analysis of the accomplishment link-volume constraintp. 55
- Table 2-4. Percentage of accomplishment FL model and No FL models – Nodes (production)p. 57
- Table 2-5. Percentage of accomplishment FL model and No FL models – Links (Volumes)p. 58
- Table 2-6. Percentage of accomplishment - Descriptivesp. 60
- Table 2-7. Quantity of solutions inside and outside the rangep. 60
- Table 3-1. TTS–Descriptive analysis of accomplishmentp. 94
- Table 3-2. Percentage Distance from the target value (ΔVa) of the flow per linkp. 98
- Table 3-3. Percentage Distance (ΔVa) Descriptive - Links (volume)p. 99
- Table 3-4. Percentage Distance from the target value (ΔOi) of the production per nodep. 99
- Table 3-5. Percentage Distance (ΔOi) Descriptive - Nodes (production)p. 100
- Table 4-1. Impact of the capacity changes on the solutionsp. 134
- Table 4-2. Impact of the cost changes on the solutionsp. 135