Simple Explanation (ELI5)

It's like when you have to buy a new phone: you need to balance price, battery life, camera, and durability. The algorithm does the same for vans, turning comments like 'high cost' into numbers and finding the best choice for the company.

Key Advantages & Current Limits

🟢 Pros: Provides an objective ranking balancing costs and environmental impact. 🟢 Pros: Handles uncertainty in human judgments using fuzzy logic. 🔴 Cons: Weight selection depends on experts and can be subjective. 🔴 Cons: The model is complex and requires technical expertise to implement.

Future Outlook & Takeaways

In the next 3-5 years, this model will be integrated into fleet management software, enabling SMEs to optimize costs and reduce emissions. Additionally, real-time data will improve decision accuracy.

Green fleet: the algorithm that picks the best vehicles?

Article

100%

Imagine having to replace all your company vans: which ones do you buy? Diesel, electric, hybrid? For a small business, it is not just about price, but emissions, range, charging, and future maintenance costs. Decisions are often made "by gut feeling", risking costly mistakes. Now mathematics comes to the rescue: researchers have developed a hybrid model combining fuzzy logic (which handles nuances, not just yes/no) with the TOPSIS method, used to compare complex alternatives.

The system works like this: experts express judgments in words ("high cost", "medium range"), which the algorithm converts into "triangular" numbers capable of representing uncertainty. It then crosses six criteria — total cost, CO2, maintenance, range, charging compatibility, energy independence — and produces an objective ranking. The result? A priority index (FRPI) that tells exactly which vehicles to scrap now and which to keep, balancing wallet and planet.

In the simulated test, the model exposed vans that looked cheap but polluted too much and cost a lot to maintain, promoting instead electric ones with a good charging network. A digital compass for SMEs and public bodies wanting to modernize without being overwhelmed by complexity. Sustainable mobility is no longer just an ideal, but a transparent calculation.

Bonus

100%

The core of the model is the hybridization between Fuzzy TOPSIS and FRPI (Fleet Renewal Priority Index). Fuzzy logic handles the inherent imprecision of human judgments: instead of crisp values, Triangular Fuzzy Numbers (TFNs) defined by a triplet (l, m, u) for lower limit, modal value, and upper limit are used. This allows modeling uncertainty on qualitative criteria like "infrastructure compatibility".

The process follows rigid steps: 1) Criteria and weight definition (via Fuzzy AHP or expert judgment). 2) Fuzzy decision matrix construction. 3) TFN normalization (linear for benefit/cost criteria). 4) Calculation of Fuzzy Positive Ideal Solution (FPIS) and Negative Ideal Solution (FNIS). 5) Distance of each alternative from FPIS/FNIS. 6) Closeness Coefficient (CCi) for TOPSIS ranking. 7) Integration with FRPI: normalized CCi weighted by operational urgency (vehicle age, mileage, historical failures).

The final output is a composite score: high values = urgent replacement with high-score TOPSIS option (e.g., BEV if infrastructure exists); low values = retention. The simulated case study on a mixed 20-vehicle fleet showed sensitivity to CO2 weight: increasing w_CO2 from 0.15 to 0.30, 3 Euro 6 diesel vans rose to top-5 replacement priority, overtaking plug-in hybrids with low electric range. It demonstrates how corporate policies (carbon budget) radically change the optimal technology strategy.