Eduardo Salazar-Treviño
Research Subjects
Eduardo Salazar-Treviño has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Al-Amin Khan
Department of Mathematics, Jahangirnagar University, Dhaka, Bangladesh
1 shared publication - Aleix Llenas
Data Science & Advanced Analytics Department, PepsiCo, Inc., Purchase, New York 10577
1 shared publication - Cynthia Griselle De-La-Cruz-Márquez1 shared publication
David Bofill
Data Science & Advanced Analytics Department, PepsiCo, Inc., Purchase, New York 10577
1 shared publicationDeepika Gowda
Data Science & Advanced Analytics Department, PepsiCo, Inc., Purchase, New York 10577
1 shared publicationFederico Todeschini
Data Science & Advanced Analytics Department, PepsiCo, Inc., Purchase, New York 10577
1 shared publication
Latest Publications
- Open Access
- 15 September 2026
- Published by Elsevier in SSRN Electronic Journal
A Flexible Iterated Local Search for the Latency Location Routing Problem with Opening Costs and Split Deliveries
The Latency Location Routing Problem with Opening Costs and Split Deliveries (LLRP-OCSD), introduced by \cite{nucamendi2026latency}, integrates facility selection, cumulative vehicle routing, and split-delivery decisions. Its latency-based objective adopts a human-centric perspective by prioritizing early service and reducing customer waiting times, making the problem relevant to time-sensitive applications such as healthcare logistics and post-disaster relief operations. Split deliveries provide operational flexibility by allowing customer demand to be allocated among multiple vehicles, especially when vehicle capacity is limited. This paper proposes a flexible Iterated Local Search (ILS) for the LLRP-OCSD. The method combines a randomized, split-aware construction procedure with a Variable Neighborhood Descent that explores intra- and inter-route neighborhoods. Diversification is achieved through randomized route exchanges, vehicle--depot reassignment, controlled acceptance of non-improving solutions, and restart mechanisms. Computational experiments on the 38 benchmark instances show that the proposed method establishes new best-known solutions (BKS) for 32 instances, improving the previous BKS by 9.66\% on average and by up to 67.32\%. The algorithm also exhibits consistent performance across independent executions, with an average coefficient of variation of 0.79\%. These results demonstrate the value of jointly exploring location, routing, and split-delivery decisions for designing responsive and operationally flexible distribution systems.
- 15 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Flexible Iterated Local Search for the Latency Location Routing Problem with Opening Costs and Split Deliveries
The Latency Location Routing Problem with Opening Costs and Split Deliveries (LLRP-OCSD), introduced by \cite{nucamendi2026latency}, integrates facility selection, cumulative vehicle routing, and split-delivery decisions. Its latency-based objective adopts a human-centric perspective by prioritizing early service and reducing customer waiting times, making the problem relevant to time-sensitive applications such as healthcare logistics and post-disaster relief operations. Split deliveries provide operational flexibility by allowing customer demand to be allocated among multiple vehicles, especially when vehicle capacity is limited. This paper proposes a flexible Iterated Local Search (ILS) for the LLRP-OCSD. The method combines a randomized, split-aware construction procedure with a Variable Neighborhood Descent that explores intra- and inter-route neighborhoods. Diversification is achieved through randomized route exchanges, vehicle--depot reassignment, controlled acceptance of non-improving solutions, and restart mechanisms. Computational experiments on the 38 benchmark instances show that the proposed method establishes new best-known solutions (BKS) for 32 instances, improving the previous BKS by 9.66\% on average and by up to 67.32\%. The algorithm also exhibits consistent performance across independent executions, with an average coefficient of variation of 0.79\%. These results demonstrate the value of jointly exploring location, routing, and split-delivery decisions for designing responsive and operationally flexible distribution systems.
- 11 June 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Applied Analytics
PepsiCo Deploys AI-Driven Pricing and Promotion Optimization at Scale
PepsiCo deployed two AI-driven systems, PromoAI and PricingAI, that combine machine learning with mathematical optimization to automate promotional calendar generation and base price recommendations at enterprise scale. Both tools have been adopted globally, compressing planning cycles from weeks to minutes while delivering measurable revenue and margin improvements.
- 11 June 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Applied Analytics
PepsiCo Deploys AI-Driven Pricing and Promotion Optimization at Scale
PepsiCo deployed two AI-driven systems, PromoAI and PricingAI, that combine machine learning with mathematical optimization to automate promotional calendar generation and base price recommendations at enterprise scale. Both tools have been adopted globally, compressing planning cycles from weeks to minutes while delivering measurable revenue and margin improvements.