Safety Resilience Evaluation of Tower-Crane Construction Using a Cloud Model and an Intuitionistic Fuzzy Bayesian Network
This study proposes a safety-resilience evaluation framework for tower-crane construction that integrates a cloud model with an intuitionistic fuzzy Bayesian network to assess system robustness under uncertainty, identifying key improvement factors through a case study application.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are the captain of a massive, floating city made of steel, where the weather changes every hour and the crew is constantly swapping out. In this world, safety isn't just about checking a box to see if a bolt is tight; it's about how well the whole ship can take a hit, bounce back, and keep sailing even when things go wrong. This is the world of "safety resilience." Think of it like a superhero's immune system: it's not just about having strong armor (that's absorptive capacity), but also how fast you can heal a wound (that's recovery capacity) and how smartly you adapt your strategy when the enemy changes tactics (that's adaptive capacity).
For years, engineers have tried to measure this resilience using simple checklists or single numbers, but those tools are like trying to describe a storm with a single raindrop. They miss the messy, uncertain, and human parts of the story. To fix this, researchers have started using "fuzzy" logic (which handles the "maybe" and "I'm not sure" parts of human thinking) and "cloud models" (which turn vague words like "pretty good" into actual numbers). Now, a new study takes these tools and combines them with a "Bayesian network," which is basically a giant, smart flowchart that connects the dots between different causes and effects, allowing us to see how a small problem in one area can ripple out to affect the whole system.
The paper you're about to read dives into the specific case of tower cranes—the giant, swan-necked machines that build our skyscrapers. The authors, Tingting Nie, Bo Liang, and Jian He, realized that while we know a lot about crane accidents, we don't have a great way to measure how resilient a crane is before a disaster happens. They built a new digital "crystal ball" to test this.
Here is what they did and found:
They started by gathering a team of five experts—senior engineers, inspectors, and professors—to look at a specific tower crane (a QTZ125 model) used in a housing project called Hongling Jiayuan. Instead of just asking the experts "Is this safe?", the researchers asked them to use descriptive words like "very high," "moderate," or "very low" to judge 12 different factors, from the crane's structural strength to how well the crew communicates.
Because human language is messy, the team used a special math trick called an "Intuitionistic Fuzzy Set." Imagine you are rating a movie. You might say, "I really liked it (support), but the ending was confusing (opposition), and I'm not entirely sure if I'd watch it again (hesitation)." This math method captures all three feelings at once, rather than forcing a simple "yes" or "no." They then fed these fuzzy ratings into their "cloud model," which turned those vague opinions into precise probabilities.
Finally, they plugged everything into their Bayesian network flowchart. This allowed them to simulate how the crane would perform under pressure. The result? The crane in the study had a comprehensive resilience score of 0.785, and there was a 70.6% probability that it was in a "high resilience" state. In plain English, the crane was doing pretty well overall. It was particularly strong at "absorbing" shocks (like having a solid foundation and good safety devices) and "adapting" to changes (like having good management).
However, the study also spotted the weak links. The crane was weakest in "recovery capacity"—meaning if something went wrong, it might take longer than ideal to fix it and get back to work. When the researchers ran a "sensitivity analysis" (a test to see which factors would cause the biggest drop in safety if they got worse), they found the top five things that needed the most attention:
- Standard updating and continuous improvement (making sure the rules keep up with reality).
- Organizational management and responsibility closure (making sure someone actually takes charge and finishes the job).
- Installation, dismantling, and jacking control (the tricky parts of putting the crane up and taking it down).
- Foundation and attachment stability (making sure the base is solid).
- Inspection, maintenance, and hazard rectification (fixing problems before they become disasters).
The authors are careful to say that this model doesn't predict accidents or replace a physical inspection. Instead, it acts like a diagnostic tool that translates human judgment and paper records into a clear, traceable map of where the system is strong and where it needs a tune-up. They suggest that by focusing on the "root causes" like management and rule-updating, rather than just fixing the immediate symptoms, construction sites can build cranes that don't just stand up, but truly bounce back.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.