Inter IIT 12 - Team_59 - Solution Quantum and Classical Pathfinding and Airline Re-accommodation
Overview The project aims at addressing the challenge of passenger re-accommodation due to planned schedule changes in airlines. The key objectives include identifying impacted flights and passengers, determining suitable alternate flights, ranking alternate flight solutions, and prioritizing PNR re-accommodation. The solution incorporates a business rule engine for flexibility in rule application and produces two sets of solution files: a default flight level re-accommodation solution and an exception list for individual PNR re-accommodation. The project emphasizes the creation of various rule profiles, flexibility in rule enforcement, and the generation of comprehensive solution files catering to different scenarios.
Files Description
Python Scripts
* classical_re-assignment_greedy.py
* find_impacted_passengers: Identifies passengers affected by flight changes.
* calc_score: Calculates scores for reassignment options.
* find_possible_paths: Determines potential paths for reassignment.
* main: The main function orchestrating the process.
* PNRS: Manages Passenger Name Records.
* classical_pathfinding.py
* Multiple functions including init, statistics_printer, reduce_network, calc_score, recurse, solver,
* multisolve, solve, find_impacted_passengers, and matrix_finder, all contributing to the classical approach to pathfinding.
* CLASSICAL_FINAL.py
* main: The final integration of classical pathfinding and reassignment methods.
* QUANTUM_FINAL.py
* In this file, we have formulated our solution into two sub problems:
* the first part focuses on path finding where we have modelled our problem into a Constrained Quadratic Model(CQM)
* The second part focuses on passenger
* re-accomodation(which we modelled as a CQM too)and generating the solution files.
* Functions like next_72hrsflights, get_direct_flights, get_1_interconnecting, get_2_interconnecting are being defined for flight data reduction.
* encode_flights, decode, and main to implement D-Wave’s HybridCQM Sampler to solve our CQM Model for the project.
* Nearby_Airports.py
* nearby_airport: Finds airports in close proximity to a given location.
* pnr_reaccomodation.py:Py file containing the below functions and classes for Flights and Passenger.
* build_knapsack_cqm: Builds a Constrained Quadratic Model for the knapsack problem involving passenger re-accommodation.
* parse_solution: Interprets solver results for re-accommodation.
* reaccomodation: Applies a CQM solver to the knapsack problem in passenger re-accommodation.
* path_scoring.py: Contains classes to score flight paths and PNRs
* Functions for scoring flights including next_72hrsflights, get_direct_flights, get_1_interconnecting, get_2_interconnecting, encode_flights, and decode_and_retrieve_flights.
Data Files
* GlobalAirportDatabase.csv
* Description: A comprehensive database of global airports, utilised as a reference in the pathfinding process.
* Other excel files given with the Dataset of the Problem Statement
* A sample output for the cancellation of flight INV-ZZ-3174758 flight using CLASSICAL_FINAL.py
* Business_Rules_PNR.csv : Has an Enabled column to turn on or off the business rules accordingly with an extra advantage of dynamic scoring for each business rule
* Flight_Scoring.csv : It is similar to the Business_Rules_PNR file, just this one is for making changes in flight scoring
Installation and Usage
docplex
qiskit_optimization
numpy
pandas
dimod
dwave
Contributing
Team_59
License
None
Github Repo Link
https://github.com/Sukhvansh2004/Mphasis-Foundation-PS-TEAM_59.git