Our Core Technologies are about Machine Learning and Algorithms. We can offer solutions for Data processing and Mining, Modeling & simulation, Decision support algorithms, Neural Networks, Optimization techniques, Fuzzy Logic Techniques, Life–Cycle Analysis (LCA), Life–Cycle Cost Analysis (LCCA).
Having acquired a labeled set of historical data, classification provides target labels to new data. The features are selected based on their discriminatory ability, avoiding overlapping, or chosen by the network itself using state of the art Deep architectures. The classification methodology can be applied to a two-class or to multi-class problems. The techniques that we use from our arsenal vary among Support Vector Machines, Multi – Layer Perceptron, Naïve Bayes, Radial Basis Function Networks, k-Neighbors Classifiers, Convolutional Neural Networks.
Use cases: Object recognition, Classification of defects on production, Face recognition, Optical character recognition, Identifying state of machinery, Inspection of Printed Circuit Boards.
Do you want to make a forecast based on a known output value? If so, regression is a prediction method, where a function is approximated to fit the data. Regression models the relationship between variables. To achieve our results we use a variety of techniques, such as Linear Ordinary Least Squares Regression, Multivariate Adaptive Regression Splines, Elastic Net, Least Absolute Shrinkage and Selection Operator, Least – Angle Regression.
Use cases: Sales growth prediction, Demand analysis prediction, Weather forecasting
How to find the optimal value among a dataset? The target of the optimization methods is to minimize or maximize a function by choosing values within certain boundaries (problem constraints) and computing the value of the function. Its problem can be formulated in a manner that it will have either just one objective or multiple ones. Some of the techniques we use are Gradient-based Methods, Genetic Algorithms, Particle Swarm Optimization, Artificial bee colony, Simulated annealing, Hill Climbing.
Use cases: Decision making in operations research, Resource utilization improvement, Cost reduction.
Modeling & Simulation
Wanting to know the outcome of a process if done in a particular way? Through our advanced modeling techniques, real life processes can be developed in a virtual environment. Our services include steady state information of a process at any chosen time, and our dynamic service offers information that describe the process appending a temporal dimension to the results.
Use Cases: applications can range from automobile suspensions, fluid & gas transportation in pipelines, to quantifying the energy savings from installing new insulation materials.
Fuzzy Cognitive Maps
If you have to deal with complex systems where their factors are characterized by dependence and feedback; if you want to integrate experts' knowledge or in case you aim to develop an automated medical assisting system or even to control a process, the chances are that you need Fuzzy Cognitive Maps (FCMs). A FCM may be perceived as a graphical representation of a system's existing knowledge, which consists of nodes and directed labeled arcs that form the nodes causal relationships. This simple, yet powerful scheme, inspired by the human cognition process, is capable of addressing problems with incomplete or missing data, quantifying human behavior or even dealing with problems where no single correct solution is available.
Use cases: medical applications (e.g. diagnosis), decision support, modeling and control, expert systems, computer assisted learning evaluation.
LCA - LCCA
Are you planning to develop a new material, product, or process? Do you want to fine-tune your product development process to maximise its sustainability? We use Life Cycle Assessment (LCA) methodologies to identify the environmental impact of innovative technologies throughout their entire lifecycle. We complement LCA with Life Cycle Costing (LCC) to bridge the environmental impacts of new technologies with their economic feasibility. The results of these studies are fed back into the product development process. As a result, we offer our clients the opportunity to integrate environmentally conscious design principles into their product development process while maximising their profit.
Use cases: Integration of sustainability in product development, optimisation of new manufacturing and production processes, environmentally conscious product design, identification of the environmental footprint of new products/technologies.