Tools to solve constraint satisfaction problems: Constraint Satisfaction Problem Solvers.
Inspired by Russell and Norvig's "Artificial Intelligence - A Modern Approach" Python code and modified under the MIT license.
Command Line:
npm i csps
app.ts:
import { csp, min_conflicts } from "csps";
// Setup your problem
const variables = ["cs108", "cs112", "cs212", "cs214"];
const possible_attributes = [
["mwf800", "nh253", "norman"],
["mwf800", "nh253", "vanderlinden"],
["mwf800", "nh253", "adams"],
["mwf800", "sb382", "norman"],
["mwf800", "sb382", "vanderlinden"],
["mwf800", "sb382", "adams"],
["mwf900", "nh253", "norman"],
["mwf900", "nh253", "vanderlinden"],
["mwf900", "nh253", "adams"],
["mwf900", "sb382", "norman"],
["mwf900", "sb382", "vanderlinden"],
["mwf900", "sb382", "adams"],
];
const domains = {
cs108: possible_attributes,
cs112: possible_attributes,
cs212: possible_attributes,
cs214: possible_attributes,
};
const neighbors = {
cs108: ["cs112", "cs212", "cs214"],
cs112: ["cs108", "cs212", "cs214"],
cs212: ["cs108", "cs112", "cs214"],
cs214: ["cs108", "cs112", "cs212"],
};
const constraints = (
class1: string,
c1Attributes: string[],
class2: string,
c2Attributes: string[],
): boolean => {
/*
Constraints for class scheduling
c1 and c2 are tuples in the form (time, room, faculty)
returns true if constraints are met.
The constraint that there is only one section of class
is implicit because classes are variables.
*/
// Return true if same class.
if (class1 === class2) {
return true;
}
// Check to make sure faculty is not teaching at the same time
if (c1Attributes[0] === c2Attributes[0] && c1Attributes[2] === c2Attributes[2]) {
return false;
}
// Check to make sure class is not in the same room at the same time
if (c1Attributes[0] === c2Attributes[0] && c1Attributes[1] === c2Attributes[1]) {
return false;
}
return true;
};
const aCSP = new CSP<string>(variables, domains, neighbors, constraints);
// run min_conflicts on problem
const res = min_conflicts(aCSP);
console.log(res);
Please feel free to create issues or make PRs. I will take a look at them and will address them as I see fit.