Freelancer profile translated to English.

Description

🎯 Have a complex problem that no one could solve? That's exactly what I do.

I am Aimen Khiar, a PhD researcher at the University of Haute-Alsace / CESI Campus of Strasbourg, specializing in mathematical optimization, AI, and complex systems modeling. What sets me apart? I'm not just a developer – I'm a scientist who solves problems at the intersection of theory and reality.

My work focuses on the planning of electric vehicle charging under stochastic uncertainty. I proposed the first multi-objective model of its kind, published in OMEGA (Q1, IF 7.2) – one of the most selective journals in management science worldwide. My research has also been presented at IEEE CEC 2026 (top 3 globally), ICORES 2025/2026, and ROADEF 2025/2026.

🛠️ Key Skills:

• Combinatorial & Stochastic Optimization: MILP, scheduling, routing, planning under uncertainty

• Multi-objective Metaheuristics: NSGA-II, NSGA-III, MOCS, MOPSO, Simulated Annealing

• Probabilistic & Statistical Modeling: Monte Carlo, probability distributions, GARCH models and extensions

• Time Series & Forecasting: modeling, prediction, anomaly detection, application to exchange rates and financial indices

• Data Analysis: exploration, visualization, cleaning, statistical interpretation

• Machine Learning & Data Science: supervised, unsupervised, federated learning

• Scientific Development: Python (NumPy, Pandas, Numba, Matplotlib), Gurobi, CPLEX, LaTeX, GitHub

💼 I solve your optimization, scheduling, routing, or data science problems with the rigor of a researcher and the efficiency of a practitioner.

🎓 Graduated top of my class — Master's in Stochastic Modeling & OR (USTHB).

💰 Daily Rate: €300 / day — Available remotely.

Industry field of expertise

Languages

French
Native or bilingual
English
Native or bilingual
Arabic
Native or bilingual

Workplace preferences

Remote only

Primarily works remotely

Université
PhD Researcher
February 2024 - Today (2 years and 6 months)
Mulhouse, France
PhD Researcher in Operations Research — University of Haute-Alsace / CESI Campus of Strasbourg

📌 The problem I solve
Charging stations face a critical challenge: how to plan the charging of multiple vehicles across several stations — who charges before whom, on which charger, at what time — while optimizing profit, peak power, lateness, and customer satisfaction, and considering uncertainty in arrivals, charging durations, and electricity prices? This problem is combinatorial, stochastic, and multi-objective.

🔬 What I've accomplished
I have proposed original models for multi-objective electric vehicle charging planning, combining advanced techniques such as Mixed Integer Programming (MILP), probabilistic modeling, and hybrid metaheuristics (NSGA-II, NSGA-III, MOCS, MOPSO).

📄 Publications & Conferences
Published in leading journals and conferences, including:

• OMEGA – Q1, IF 7.2 — doi.org/10.1016/j.omega.2025.103506
→ First multi-objective model for charging planning under stochastic durations: scheduling across multiple chargers with uncertain service times, simultaneously minimizing average lateness and peak load, while maximizing the amount of energy delivered to customers.

• IEEE CEC 2026 — WCCI (Top 3 globally, Maastricht, June 2026) — accepted
→ Extension of the OMEGA model with stochastic vehicle arrivals and random demand cancellations.

• ICORES 2025 — Porto — doi.org/10.5220/0013236400003893

• ICORES 2026 — Marbella — doi.org/10.5220/0014305800004055

• ROADEF 2025 & 2026 — French National Operations Research Conference
Scheduling Stochastic Optimization Python Algorithms Mathematical Modeling
Université
Master's in Stochastic Modeling and Forecasting in Operations Research
September 2021 - July 2023 (1 year and 10 months)
Algiers, Algeria
🎓 Graduated top of class — with honors.

Final year project focused on an original extension of the GARCH model for modeling the volatility of financial series. I proposed the BPGARCH (Buffered Periodic GARCH) model, which incorporates a periodic structure and a buffering mechanism to better capture the non-linear dynamics of financial markets. The results show that BPGARCH outperforms the classic GARCH and several of its recognized extensions in the literature, in modeling multiple stock market indices.
Time Series Time Series Forecasting Data Analysis Python