Every now and then I build a project simply because I want the tool to exist, this was one of those projects. After Frontier introduced Colonisation into Elite Dangerous, I found myself repeatedly asking the same questions:
- Which facilities do I actually need?
- What should I build next?
- How can I maximise my system’s economy?
- Is there an optimal build order that avoids wasting construction points?
There were already some excellent planning tools available, but they all focused on manually designing systems. What I was missing was something that could answer those questions automatically.
Introducing EDCPS
The Elite Dangerous Colonisation Planner & Solver (EDCPS) is an open-source web application that can both plan a colonised system manually and automatically generate an optimal build plan based on your goals. Instead of experimenting with countless facility combinations, you simply describe your current system, specify what you’re trying to achieve, and let the solver do the hard work.
The project is available here: https://gaborauth.github.io/ed-colonisation-planner
Source code also available: https://github.com/gaborauth/ed-colonisation-planner/
From Planner to Solver
EDCPS started as a visual planner, but quickly evolved into something much more interesting. At its core is a deterministic optimisation engine that analyses possible facility layouts while considering:
- facility dependencies
- construction point requirements
- orbital and surface slot limitations
- body restrictions
- increasing construction costs
- economy interactions
- user-defined objectives and constraints
The result isn’t just a list of facilities to build – it also produces the recommended construction order, ensuring prerequisites are met while working towards your target economy or score. Given the same input, the solver will always produce the same solution. There is no randomness involved.
Different Goals, Different Solutions
Not every commander wants the same colony. Some players simply want the highest Wealth score. Others are trying to unlock a specific economy, maximise Industrial output, balance several economy types, or work around the facilities they’ve already constructed. The solver supports these different objectives instead of assuming there’s only one “correct” answer. Some examples include:
- finding the minimum facilities required to unlock a target economy
- maximising a particular economy score
- balancing multiple economy types
- continuing from an already partially built system
- respecting body and slot constraints
- generating an efficient build order
The tutorials cover several practical examples and explain how different optimisation goals affect the resulting plan.
Built Around Real Gameplay
One of the design goals was to integrate naturally with existing Elite Dangerous tools. EDCPS can:
- import Journal files
- import systems directly from Spansh
- edit plans manually
- save multiple plans locally
- experiment with different optimisation strategies
Almost everything runs directly in your browser. Your journal files and saved plans remain on your own machine, with only optional lookups sent to external services.
Behind the Scenes
One misconception I’ve already seen is that the solver must be using AI. It doesn’t. The solver is built entirely on deterministic algorithms and traditional optimisation techniques. The challenge wasn’t generating plausible answers—it was efficiently exploring a huge search space while respecting all the dependency rules, construction costs, slot limitations and economy mechanics.
Open Source
Like many of my side projects, EDCPS is completely open source.
That means anyone can inspect the source code, report issues, suggest improvements or contribute new features. As Frontier expands the colonisation mechanics, I expect the planner to evolve alongside the game.
A Surprisingly Interesting Problem
From a software engineering perspective, this turned out to be one of the most enjoyable projects I’ve worked on recently. Colonisation planning looks simple at first, but every decision affects future possibilities. Building one facility may unlock several others, increase construction costs, consume limited slots, or influence multiple economy scores. The number of possible combinations grows extremely quickly, making brute-force search impractical. Finding algorithms that could consistently produce good solutions within a few seconds became an interesting optimisation problem in its own right.
Give It a Try
If you’re playing with the new colonisation mechanics, I’d love to hear how well EDCPS works for your systems.
Bug reports, feature requests and pull requests are always welcome.
Fly safe, CMDRs.