Asset Publisher

Host Research Group

ES27_MATHMODE _David Pardo Zubiaur / Michael Barton

David Pardo (MATHMODE at UPV/EHU + MATHDES at BCAM) y Michael Barton (MATHDES at BCAM)


Group description

Our research spans areas of Deep Learning, Inverse Problems, Finite Elements, Massive Computations, Numerical Analysis, Geometric Modeling, Computer Aided Design, and Modeling of Manufacturing Processes. We work in close interaction with industrial partners and institutions to promote transfer of knowledge and obtain feedback from real-life applications. In relation with ADMIRE, we are currently collaborating with the following research projects:

Geometric Modeling with Manufacturing Applications: We model manufacturing processes such as 5-axis Computer Numerically Controlled (CNC) machining, where we design and implement path-planning algorithms to navigate the CNC machine such that the error between the surface and the tool is minimized. We compute highly-accurate motions of on-market tools, as well as look for optimal shape of the tool to comfort the given input free-form geometry such as impellers, blisks, and/or turbine blades (MACROPISTAS project).

Microstructural design of free-form objects: We Analyse, Design, and Manufacture objects using Microstructures (ADAM^2). The evolution of new manufacturing technologies such as multi-material 3D printers gives rise to new types of objects that may consist of considerably less, yet heterogeneous, material, consequently being porous, lighter and cheaper, while having the very same functionality as the original object when manufactured from one single solid material.


  • 5-axis CNC machining
  • Computer-aided manufacturing
  • Curvature adapted machining
  • Path and motion planning
  • free-form surface
  • Custom-shaped milling tool
  • Microstructural design

Team Description

  • David Pardo (Principal Investigator)

    ORCID: 0000-0002-1101-2248

  • Michael Barton (Co-Principal Investigator)

    ORCID: 0000-0002-1843-251X

  • Abdessamad Ousaadane (Post-Doctoral Researcher)

    ORCID: 0000-0001-6672-7745

  • David Rochera (Post-Doctoral Researcher)

  • Ali Hashemian (Post-Doctoral Researcher)

    ORCID: 0000-0002-8190-222X


  • ADAM^2

    Pl: M. Barton

    Funding Agency*: EU

    Ongoing: yes

    Project reference: 862025 - ADAM^2 (H2020)


    Pl: M. Barton

    Funding Agency*: EU

    Ongoing: no

    Project reference: 764979 - ENABLE (H2020)


    Pl: M. Barton

    Funding Agency*: NAT

    Ongoing: yes

    Project reference: PID2019-104488RB-I00


    Pl: M. Barton

    Funding Agency*: RE

    Ongoing: no

    Project reference: KK-2020/00102

  • Ramon y Cajal

    Pl: M. Barton

    Funding Agency*: NAT

    Ongoing: yes

    Project reference: RYC-2017-22649

* INT - International EU - European NAT - National RE - Regional


Research Lines


Feedrate optimization for 5-axis flank CNC machining

  • Energy consumption of a 5-axis machining center is affected by higher order kinematic terms (acceleration, jerk) and we aim to reduce it by optimizing the tool kinematics. Recently, we have shown that one can considerably reduce the jerk in the context 5-axis flank CNC machining by applying a proper metric that measures distance between lines in 3D space. However, this reduction was achieved with respect to the workpiece coordinates system. We aim to adapt the so called ruled-distance metric to the machine coordinate system, and consequently minimize the energy consumption of the milling process.


Design and optimization of hollow milling tools

  • As recently shown in the work of the PI, higher order differential analysis of contact between the tool and the workpiece give rise to highly accurate path-planning algorithms that outperform the state-of-the-art CAM software. Even higher accuracy is possible if the tool is also considered as an unknown. Some curved tools, however, may not be the best in terms of fluid (coolant) flow around the tool and we aim to look for hollow tools that would admit the coolant pass throughout the cutting tool. This would result in smaller tool wear and higher sustainability of the whole manufacturing process.

(5+1)-axis Manufacturing with dynamical cutters

  • Traditionally, the milling tool is static, however, the current technological developments allow us to design tools that are dynamical, i.e., tools that change their shapes during the machining process. One such a tool is an umbrella-like mechanism which has one degree of freedom to dynamically adapt the opening angle, of basically a conical tool, to the geometry of the workpiece. We aim to design path-planning algorithms to navigate these types of milling tools. Conceptually, this research will correspond to designing curves in a 6-dimensional construction space, and will go towards modern, “(5+1)-axis” CNC machining.

Cross-border Collaboration (if any)

Our group has a well-established collaboration with the High Performance Machining Group, led by Prof. Norberto de Lacalle, at UPV. We have several past projects sealed by Q1 publications (see some in the table above), and several ongoing ones. In particular, we currently work on path-planning problems where the milling tool itself is a variable that we compute in our optimization-based framework.

A potential collaboration can also arise with the group of Olivier Cahuc (University of Bordeaux). Dr. Cahuc is coordinating the EU funded ETN-ITN project (ENABLE) and M.Barton is one of the PIs.