Question 1

What is operations research?

Accepted Answer

Operations research is the discipline that applies mathematical modeling, statistics, and optimization to make better decisions in operational problems. It includes mathematical programming, discrete-event simulation, queueing theory, and metaheuristics.

Question 2

What is the difference between optimization and machine learning?

Accepted Answer

Machine learning learns patterns from historical data and answers the question "what will happen". Optimization finds the best decision given an objective and constraints, and answers "what to do". The two complement each other: many systems use ML to predict demand, prices, or risk, and optimization to allocate resources or schedule operations based on those predictions.

Question 3

What is stochastic simulation?

Accepted Answer

Stochastic simulation models systems with uncertainty through thousands of random scenarios to estimate their behavior. It is used to evaluate operational policies before implementing them: how many trucks to operate, what inventory level to maintain, what shifts to assign. Unlike optimization, it does not search for an optimal solution directly, but rather measures the expected performance and variability of different alternatives.

Question 4

What is a mathematical optimization model?

Accepted Answer

An optimization model formally describes a decision: an objective function (what to maximize or minimize, like cost or profit), decision variables (what to choose, like quantities, routes, or assignments), and constraints (physical, contractual, or capacity limits). Solving it produces the combination of decisions that respects the constraints and optimizes the objective. The most common techniques include linear programming, mixed-integer programming (MIP), nonlinear programming, and heuristic methods.

Question 5

What is business intelligence and how is it different from machine learning?

Accepted Answer

Business intelligence (BI) organizes historical data into reports, dashboards, and KPIs for human decision support. Machine learning builds predictive models from that same data. BI answers "what happened"; ML answers "what will happen". Both rely on clean, well-modeled data, but BI ends in visualizations for people, and ML ends in models that automate decisions or feed other systems (including optimization).

Question 6

How long does a typical Lontra project take?

Accepted Answer

Project duration depends on scope, but engagements are structured in defined phases: an initial consultation of 1 to 3 weeks, a modeling and validation phase of 4 to 12 weeks, and an implementation phase whose duration depends on the client's systems. Complete projects typically run between 2 and 6 months.

Question 7

What does Lontra deliver at the end of a project?

Accepted Answer

Every project ends with a concrete deliverable: a working optimization model, a simulator, a decision-support system, or a set of predictive models integrated with the client's systems. The handover includes documented code, validation against real client data, and a transfer session with the client's internal team.

Where Mathematics
Meets Momentum.

Our Mission

Our Methodology

1. Data Synthesis

2. Algorithmic Rigor

3. Decision Impact

Meet the Team

Juan Pablo Morande

school Education

work Selected Applied Projects

Collaborator Network

Konstantinos Ziliaskopoulos

Jose Palma Jofré

Have an operational
challenge?

Where Mathematics Meets Momentum.