DyMDeST ? Dynamic Multi-criteria Decision Support Tool

Ref-Nr: TDO0145

Technology abstract

Developed in Portugal, by a Research and Development Institute, the offer is an algorithm providing an intelligent decision support tool that is capable of evaluating multiple heterogeneous input data (criteria) and dynamically adapting to changes in the system and able to handle uncertainties in the data.

Having being developed and used in the space domain, we seek a partnership with entities from other domains (e.g. insurance, energy, health) to apply this in their respective fields.

This technology has received various expression of interest and reveals an application potential that involves decision making with multiple criteria, namely for Manufacturing/industrial, Medical/health, Banks/financial institutions. A feasibility study sponsored by the ESA Tech Transfer Programme Office has been performed by UNINOVA and demonstrates that this algorithm is also useful for intelligent landing of Unmanned Aerial Vehicles (UAVs): http://ptti.ipn.pt/news_items/30.

- -

Read more about this broker

Technology Description

The algorithm was originally developed as a method for selecting the best landing site for spacecraft, based on criteria such as terrain texture, slope, available fuel, reachability, shadow etc. It has since been applied in a technology transfer to Unmanned Aerial Vehicles (in partnership with an SME).

The algorithm is an intelligent mathematical model capable of handling uncertainty in dynamically changing heterogeneous input data and criteria, during a spatio-temporal decision process that considers current and historic information to achieve a conclusion (consensus).

In essence it is able to evaluate a large set of criteria, adapting the weights according to changes over time in order to provide the optimal option at any given moment.

It can be readily applied to any domain where a set of criteria have to be weighted by relative importance and evaluated to produce an optimal solution.

Innovations & Advantages

The novelty of the multi-criterion decision method is its dynamic characteristic and capacity to deal with dynamically changing input data and criteria. Further, it involves a combination of specialized soft computing concepts and techniques, from specialized reinforcement operators, fuzzy normalization and heuristic optimization algorithms to efficiently search the solution domain.

In time constrained (time critical) applications it is also possible to engage a ?non-exhaustive? search optimization process, which can provide faster-than-real-time solutions (albeit at the expense of slight decrease in guaranteeing an optimal solution).

 

Current and Potential Domains of Application

  • Energy sector
  • Environment
  • Bioinformatics
  • Insurance sector