A parallel research thread running underneath the industrial-automation career: two decades of peer-reviewed work in process intensification and computational optimization, done as an industry collaborator alongside university research groups in Mexico — first through Innovación Integral de Sistemas (2002–2009), then through Exxerpro Solutions from 2010 onward. This is the same first-principles, evidence-based reasoning applied to a different class of problem — reducing energy consumption and cost in distillation-based separation processes using mathematical modeling, evolutionary algorithms and, more recently, neural networks as surrogate models.
Citation metrics fetched directly from the live Google Scholar profile (name "Abel Briones," affiliation "Exxerpro Solutions SA de CV") on 2026-09-15 — they will keep climbing. Publication count reflects individually opened and verified PDFs, not a claimed total; the Scholar profile may list additional items not yet collected here.
Two intertwined lines of work, almost all produced with the same core group of collaborators at Universidad de Guanajuato, CIATEQ and the Universidad Autónoma de Querétaro.
Design and control analysis of dividing-wall columns, Petlyuk sequences and other thermally coupled arrangements that cut distillation's energy and capital cost — the single largest body of work, and the most-cited.
A multi-objective genetic algorithm with restrictions, coupled to the Aspen Plus process simulator, used across most of these papers as the core design and optimization tool — one of the more cited applications of GA-based design in this subfield.
Using artificial neural networks to approximate expensive objective-function evaluations and speed up the genetic algorithm itself — an early, direct precursor to the applied-AI work now on the industrial side of the portfolio.
Process intensification and energy integration for producing renewable jet fuel and green diesel from Jatropha curcas oil via hydrotreating — applying the same optimization toolkit to a sustainability-driven separation problem.
19 peer-reviewed journal papers, newest first. Each PDF was opened individually to confirm the real title, authors, venue and DOI — several corrected DOIs that appeared in earlier internal notes (they had been guessed rather than read off the document). Descriptions below are written for this page, not copied from the papers' own abstracts. Citation counts shown are the live per-paper figures from Google Scholar where available.
A practical procedure — with Abel as lead author — for wiring MATLAB's optimization routines directly to Aspen Plus so any multi-objective strategy can drive a full process-simulator model, instead of being limited to simplified reduced models.
doi.org/10.13053/CyS-22-4-3087 ↗Combines energy integration with process intensification in a Jatropha-curcas-to-biojet hydrotreating route, showing utility demand and CO₂ emissions both drop without materially raising production cost.
doi.org/10.1016/j.cep.2016.10.007 ↗Compares the control properties of conventional and thermally coupled distillation trains used to fractionate renewable bio-jet fuel and green diesel — asking which configuration is easiest to actually operate, not just cheapest to build.
doi.org/10.1002/ceat.201600095 ↗Redesigns the purification stage of biojet-fuel production around thermally coupled distillation plus a turbine that recovers energy from the reactor effluent, cutting the separation stage's own energy footprint.
doi.org/10.1016/j.cep.2014.12.002 ↗Tests whether optimizing distillation columns directly against a rigorous non-equilibrium (rate-based) tray model — instead of the usual equilibrium-plus-efficiency shortcut — actually changes the optimal design, and when the simpler model is good enough.
doi.org/10.1016/j.cep.2014.11.001 ↗Compares four non-catalytic supercritical biodiesel routes on energy use, cost and CO₂ footprint, and finds no single process wins on every metric at once.
doi.org/10.1007/s10098-015-0933-x ↗Swaps in a neural-network surrogate for the expensive parts of a genetic-algorithm evaluation loop when optimizing dividing-wall columns — a direct, early precursor to the applied-AI work on the industrial side of this portfolio.
doi.org/10.15255/CABEQ.2014.2132 ↗Shows that the numerical method used to fit a thermodynamic model's own parameters — not just the model itself — measurably changes the optimal azeotropic-column design, and argues for global rather than local parameter-fitting.
doi.org/10.1021/ie4019885 ↗Compares the open-loop dynamic behavior of single- and double-dividing-wall columns, to check whether the extra energy savings of the double-wall design come at the cost of harder control.
doi.org/10.1021/ie401332p ↗Optimizes a hybrid distillation/melt-crystallization process for separating close-boiling isomers, and finds the thermally coupled version of the hybrid holds up well on both cost and control properties.
doi.org/10.1016/j.cep.2012.11.007 ↗A related but separately published treatment of the same hybrid distillation/crystallization optimization problem — same team, same core method, extended into a second venue. No DOI is printed in the source PDF; verify directly against the journal or Scholar before citing.
Looks for quaternary-mixture distillation configurations that use fewer than the conventional N−1 columns, finding design tendencies driven mainly by mixture and feed composition rather than a single universal shortcut.
doi.org/10.1016/j.cherd.2012.02.004 ↗Extends the dividing-wall-column design method to columns with two dividing walls, finding total annual cost differences between one- and two-wall designs come down mainly to energy requirements.
doi.org/10.1002/ceat.201100176 ↗Applies the group's genetic-algorithm design tool to reactive distillation with thermal coupling (fatty-ester production as the case study), mapping the full cost/energy trade-off rather than a single point design.
doi.org/10.1021/ie101290t ↗Runs the constrained multi-objective GA against a rigorous Petlyuk-sequence design model and finds more energy-efficient designs than earlier reported structures, including some with four interconnecting stages instead of the usual two.
doi.org/10.1016/j.compchemeng.2010.10.007 ↗The group's most-cited paper: applies the constrained genetic-algorithm method to extractive dividing-wall columns, showing extractive separations are feasible in a single intensified shell rather than needing separate columns.
doi.org/10.1021/ie9006936 ↗Introduces the core constrained multi-objective GA, coupled to Aspen Plus, used throughout this whole body of work — built specifically to trace the full Pareto front of Petlyuk-sequence designs rather than a single optimum.
doi.org/10.1016/j.compchemeng.2008.11.004 ↗Extends the GA design tool from ternary to quaternary mixtures, testing the effect of relative feed volatilities on energy, cost and control properties of the intensified sequences.
doi.org/10.1016/j.compchemeng.2009.04.011 ↗One of the earliest papers in the series: establishes the genetic-algorithm-plus-Aspen-Plus method for designing single- and double-dividing-wall columns and testing their theoretical control properties. The DOI printed on the paper itself corrects an earlier, incorrectly transcribed one.
doi.org/10.1002/ceat.200800116 ↗7 further citable works, published as chapters in Elsevier's Computer Aided Chemical Engineering series (ESCAPE symposium proceedings) — each has its own resolvable DOI, verified the same way as the journal papers above.
Builds an energy-consumption map across five quaternary feed compositions and four column configurations, extending an approach previously available only for ternary mixtures.
doi.org/10.1016/B978-0-444-63428-3.50025-4 ↗Moves past pure process optimization into the mechanical side — CFD-simulating actual sieve-tray hydraulics inside a dividing-wall column to check for flooding and flow malfunctions the process model alone wouldn't catch.
doi.org/10.1016/B978-0-444-63455-9.50064-7 ↗An early model of the UOP Honeywell-style biojet fuel route from vegetable oil, optimizing the purification stage with the group's genetic algorithm to show a high conversion of castor oil to biojet fuel and green diesel.
doi.org/10.1016/B978-0-444-63234-0.50003-8 ↗Applies dividing-wall columns — used at the time mainly on ideal mixtures — to a realistic petrochemical refinery-cut mixture, comparing whole trains of columns rather than a single unit.
doi.org/10.1016/B978-0-444-59520-1.50007-5 ↗The earliest paper in this collection to use neural networks — as approximations that stand in for expensive objective-function evaluations — to cut both the number of evaluations and the wall-clock time needed to reach the Pareto front.
doi.org/10.1016/S1570-7946(10)28066-5 ↗With Abel as lead author: a feasibility check based on material balances, followed by rigorous GA optimization, for using dividing-wall columns on azeotropic mixtures of industrial importance — reporting up to 50% energy savings versus conventional sequences.
doi.org/10.1016/S1570-7946(09)70093-8 ↗The conference-length precursor of the 2009 quaternary-mixtures journal paper above, applying the same GA/Aspen Plus method to intensified quaternary distillation systems.
doi.org/10.1016/S1570-7946(09)70091-4 ↗A further 7 items exist in the source collection as earlier-stage conference presentations (ESAT 2009, PRES-style, and a 2008 manuscript) that appear to be precursor or working-paper versions of papers already listed above, sharing the same authors and near-identical titles/abstracts. None had an independently resolvable DOI printed in the document, so they aren't individually linked here to avoid guessing a citation; they're kept on file and can be added if Abel wants them itemized separately.
Almost all of this work was produced with the same core group of collaborators across three institutions.
CIATEQ, A.C. / Universidad Autónoma de Querétaro — the most frequent co-author across this entire collection, including several two-author papers with Abel.
Universidad de Guanajuato — corresponding author on many of the group's papers and the academic lead for most of the graduate-student-authored work.
Universidad de Guanajuato — first author on the dividing-wall-column design and control papers, one of the most frequent collaborators.
Universidad de Guanajuato — appears on the large majority of the group's distillation-design papers.
Instituto Tecnológico de Aguascalientes — thermodynamic-modeling and parameter-estimation collaborations, including the group's most-cited paper.
Instituto Tecnológico de Celaya — co-author on the Petlyuk-sequence optimization work.
Google Scholar: scholar.google.com/citations?user=pZuco48AAAAJ — live, verified.
No ORCID iD, Scopus Author ID or Web of Science ResearcherID was found in a search at the time this page was built. If Abel has one of these, it can be added directly rather than guessed at.