The Engine Behind
Intelligent Scheduling
Genera Systems is built on mathematical optimisation, not generative AI. The platform combines dependency logic, graph-based modelling, and genetic algorithms to improve construction schedules while preserving real project constraints.
Core Technology
Genera Systems uses a hybrid optimisation architecture designed for complex construction schedules. It is being developed to combine mathematical rigor with practical site logic, so the output is not only faster, but structurally sound, buildable, and relevant to real project delivery.
The long-term goal is a Construction Decision Optimisation Engine capable of optimising time, risk, resource behaviour, mitigation strategies, and schedule stability inside one connected system.
Optimisation Pipeline
Project Data
Tasks, durations, links, logic types, calendars, and planning constraints.
Dependency Graph
Activities are structured into a graph that preserves FS, SS, FF, SF logic and lags.
Optimisation Engine
The model evaluates possible sequencing and timing alternatives under constraints.
Genetic Iterations
Candidate schedules evolve through fitness evaluation, selection, crossover, and mutation.
Optimised Output
A refined schedule is produced with measurable improvement against the baseline.
Genetic Algorithms
Explore large numbers of schedule permutations to reduce project duration while preserving logic.
Graph-Based Logic
Models dependencies and relationship types so the schedule remains mathematically coherent.
Constraint Handling
Applies practical rules such as task relationships, sequencing limits, and future resource logic.
Stability Thinking
Designed to reduce unnecessary disruption when optimising schedules, not just force random movement.
Built for Real Construction Planning
- Preserves real scheduling logic and dependency structures
- Built around mathematical optimisation rather than AI guesswork
- Designed for large schedules and long planning horizons
- Structured for future resource, risk, and mitigation expansion
- Developed with construction-specific planning realities in mind
- Supports the long-term direction toward BIM-connected decision optimisation
Current Development Stage
The engine now runs inside GeneraGantt on real programmes, tested on a live 9,751-activity schedule. It returns a front of priced plans, from fastest to cheapest, rather than a single answer.
Alongside it sit DCMA 14-point checks, Monte Carlo forecasting, a risk register, progress and earned value, and a finance view. See all GeneraGantt features.