{MODELLING AND SIMULATION FOR MARKETING & MANAGEMENT}


Use of agent-based modeling, complex systems, and social network analysis to describe, explain, and predict emergent phenomena in areas such as management and marketing. Use of soft computing and artificial intelligence for validating these complex models.

Main publications:

Agent-based decision support systems for managing word-of-mouth programs. Application to freemium business model M. Chica, W. Rand. Journal of Marketing Research, 2016. More information about this line

Identimod: modeling and managing brand value using soft computing. M. Chica, O. Cordón, S. Damas, V. Iglesias, J. Mingot – Decision Support Systems 89 41-55, 2016.

Multimodal optimization: an effective framework for model calibration. M. Chica, J. Barranquero, T. Kadjanowicz, O. Cordón, S. Damas – Information Sciences 375 79-97, 2017. More information about this line

Successful projects:

ZIO (aka Identimod) is a brand management software where consultants can model a brand system (companies, organizations, cities or personal branding), run what-if scenarios to understand the behaviour of different policies, and run optimization algorithms to find the best strategies. We used system dynamics and agent-based modeling together with soft computing to model the system and perform analysis.

Identimod: Decision support system based on computational intelligence and agent based modelling for branding and marketing. See ROD company website.

{DATA SCIENCE AND MACHINE LEARNING FOR ENGINEERING AND MEDICAL APPLICATIONS}


Development and application of machine learning and data science algorithms for applications such as image recognition, pose estimation, microscopy, or wind energy forecast. Techniques such as artificial neural networks, random forest, and fuzzy rule based classification have been used by Manuel in the last years.

Main publications:

EPOSBED intelligent bed system (Patents O2011/144767, US20130152308, EP2583652A1).

Discernment of bee pollen loads using computer vision and one-class classification techniques. M. Chica, P. Campoy – Journal of Food Engineering, 112 (1-2) 50 – 59, 2012.

Successful projects:

Within EPOSBED project (FP7 for SMEs) we developed a specialty bed with an easy-to-use positioning based on Artificial Intelligence and Neural Networks to allow patients with limited mobility to change their position in bed without need of assistance from hospital personnel.

The main idea of APIFRESH (FP7 for SMEs) is to help European Beekeeping Sector by developing an automated decision support system based on image processing and artificial intelligence to discover bee-products fraud.

TOXDTECT: predicting meat life by using sensors and data analysis in food packaging.

WIND FORECAST: prediction of the global power generation of wind farms based on soft computing predictive techniques. In this project, data analysis was carried out for correcting the energy forecast of the wind farms of EDPR in Spain.  The system was installed and run real-time, improving the accuracy of the existing forecasting model.

The FORFIRE project (FP7 for SMEs) developed a new system for early detection of fires with nearly zero rate of false alarms. It will also allow for accurate localisation of the incipient fire. I developed the image processing and pattern recognition algorithms for detecting fire spots in a photodetector image.

{ OPTIMIZATION AND OPERATIONAL RESEARCH }


Mainly, topics related with multicriteria real-world problems such as multiobjective metaheuristics (evolutionary algorithms, ant colony optimization, memetic algorithms), use of preferences during and before starting the search for the optimal solutions, robust solutions in optimization, and visualization mechanisms for representing the solutions for a multiobjective problem.

Main publications: 

A multiobjective model and evolutionary algorithms for robust time and space assembly line balancing under uncertain demand. M. Chica, O. Cordón, S. Damas, J. Bautista – Omega 58, 55-68, 2016.

Multi-objective constructive heuristics for the 1/3 variant of the time and space assembly line balancing problem: ACO and random greedy search. M. Chica, O. Cordón, S. Damas, J. Bautista – Information Sciences 180 (18) 3465-3487, 2010.

Successful projects:

The project called Planning and optimization of production lines was developed by the ECSC for Tenneco Automotive. I was working with metaheuristics and production management techniques to develop a decision support system for the direction of the Tenneco plant of Gijón (Asturias).