How do you incorporate uncertainty and risk factors into your transportation network models?
由人工智能和领英社区提供技术支持
The first step is to identify and classify the sources and types of uncertainty and risk that affect your transportation network model. Some common sources are data quality, model parameters, user preferences, network conditions, external events, and policy scenarios. Some common types are aleatory, epistemic, and strategic uncertainty, and operational, tactical, and strategic risk. By defining the sources and types of uncertainty and risk, you can determine the appropriate methods and measures to deal with them in your model.
-
This is a lot of pretty words with no meat on the bone. It looks like some AI-generated text, which may work for an introduction paragraphe as the one above, but as some point you need to connect it with practical examples that are topic specific. In the present case the data sources would include census data, passenger origin-destination (O/D) surveys, mobile device data which could be generic data or data from the agency-designed app. The latter would provide the benefit of having current transit users but would ignore potential new riders. Etc... Etc...
The next step is to choose the suitable modeling framework and approach for your transportation network model. Depending on the level of detail, complexity, and flexibility that you need, you can use different types of models, such as static or dynamic, deterministic or stochastic, discrete or continuous, and linear or nonlinear. You can also use different approaches to incorporate uncertainty and risk into your model, such as sensitivity analysis, scenario analysis, robust optimization, stochastic optimization, or risk-based optimization. Each framework and approach has its own advantages and limitations, so you should consider the trade-offs and assumptions involved in your choice.
-
Again, too theoretical -- while correct, the above is at best for an academic presentation to introduce a class on the different types of modules. By itself, this paragraph offer nothing of substance which can be put to use. Either the reader knows about all the models and this paragraph is useless, or the reader wants to learn about this and essentially learns nothing.
The third step is to select the relevant performance indicators and criteria for your transportation network model. These are the measures that you use to evaluate and compare the outcomes of your model under different conditions and scenarios. Some common performance indicators are travel time, travel cost, reliability, accessibility, safety, emissions, and equity. Some common criteria are efficiency, effectiveness, robustness, resilience, and sustainability. You should choose the performance indicators and criteria that reflect your objectives and constraints, as well as the uncertainty and risk factors that you want to account for in your model.
The fourth step is to apply the appropriate methods and tools for analysis and evaluation of your transportation network model. These are the techniques that you use to generate, simulate, and optimize the solutions of your model under uncertainty and risk. Some common methods are Monte Carlo simulation, Latin hypercube sampling, design of experiments, surrogate modeling, metaheuristics, and machine learning. Some common tools are software packages, programming languages, databases, and visualization tools. You should use the methods and tools that are compatible with your modeling framework and approach, as well as the performance indicators and criteria that you selected.
-
VisionEval is a great tool (that's free and developed for R) for evaluating deep uncertainty across hundreds or thousands of scenario combinations.
The final step is to communicate the results and recommendations of your transportation network model clearly and transparently to your stakeholders and decision-makers. This is the stage where you present and explain the findings, insights, and implications of your model under uncertainty and risk. You should use clear and concise language, graphs, tables, maps, and charts to illustrate the results and recommendations. You should also highlight the uncertainties and risks involved in your model, as well as the assumptions and limitations of your analysis. By communicating the results and recommendations clearly and transparently, you can increase the credibility and usability of your transportation network model.
更多相关阅读内容
-
Transportation PlanningHere's how you can establish industry standards for performance evaluation in transportation planning.
-
Transportation PlanningHere's how you can stand out as a transportation planner: essential skills and qualifications.
-
Transportation PlanningHere's how you can enhance your career in Transportation Planning with problem-solving skills.
-
Transportation PlanningYou’re a Transportation Planning professional. How can you collect data more efficiently?