Order acceptance with reinforcement learning

M. Mainegra Hing, Aart van Harten, Peter Schuur

Research output: Book/ReportReportOther research output

21 Downloads (Pure)

Abstract

Order Acceptance (OA) is one of the main functions in a business control framework. Basically, OA involves for each order a 0/1 (i.e., reject/accept) decision. Always accepting an order when capacity is available could unable the system to accept more convenient orders in the future. Another important aspect is the aV'(tiiability of information to the decisionmaker. We use a stochastic modeling approach using Markov decision theory and learning methods from Artificial Intelligence techniques in order to deal with uncertainty and long-term decisions in Ok Reinforcement Learning (RL) is a quite new approach that already combines this idea of modeling and solution method. Here we report on RL-solutions for some OA models.
Original languageUndefined
Place of PublicationEnschede, the Netherlands
PublisherUniversity of Twente, Research School for Operations Management and Logistics (BETA)
Number of pages44
Publication statusPublished - 2001

Publication series

NameBETA working paper
PublisherUniversity of Enschede, BETA
No.WP-66

Keywords

  • IR-95693

Cite this

Mainegra Hing, M., van Harten, A., & Schuur, P. (2001). Order acceptance with reinforcement learning. (BETA working paper; No. WP-66). Enschede, the Netherlands: University of Twente, Research School for Operations Management and Logistics (BETA).
Mainegra Hing, M. ; van Harten, Aart ; Schuur, Peter. / Order acceptance with reinforcement learning. Enschede, the Netherlands : University of Twente, Research School for Operations Management and Logistics (BETA), 2001. 44 p. (BETA working paper; WP-66).
@book{f9285b84472e4b5daeb7effa4005aea0,
title = "Order acceptance with reinforcement learning",
abstract = "Order Acceptance (OA) is one of the main functions in a business control framework. Basically, OA involves for each order a 0/1 (i.e., reject/accept) decision. Always accepting an order when capacity is available could unable the system to accept more convenient orders in the future. Another important aspect is the aV'(tiiability of information to the decisionmaker. We use a stochastic modeling approach using Markov decision theory and learning methods from Artificial Intelligence techniques in order to deal with uncertainty and long-term decisions in Ok Reinforcement Learning (RL) is a quite new approach that already combines this idea of modeling and solution method. Here we report on RL-solutions for some OA models.",
keywords = "IR-95693",
author = "{Mainegra Hing}, M. and {van Harten}, Aart and Peter Schuur",
year = "2001",
language = "Undefined",
series = "BETA working paper",
publisher = "University of Twente, Research School for Operations Management and Logistics (BETA)",
number = "WP-66",
address = "Netherlands",

}

Mainegra Hing, M, van Harten, A & Schuur, P 2001, Order acceptance with reinforcement learning. BETA working paper, no. WP-66, University of Twente, Research School for Operations Management and Logistics (BETA), Enschede, the Netherlands.

Order acceptance with reinforcement learning. / Mainegra Hing, M.; van Harten, Aart; Schuur, Peter.

Enschede, the Netherlands : University of Twente, Research School for Operations Management and Logistics (BETA), 2001. 44 p. (BETA working paper; No. WP-66).

Research output: Book/ReportReportOther research output

TY - BOOK

T1 - Order acceptance with reinforcement learning

AU - Mainegra Hing, M.

AU - van Harten, Aart

AU - Schuur, Peter

PY - 2001

Y1 - 2001

N2 - Order Acceptance (OA) is one of the main functions in a business control framework. Basically, OA involves for each order a 0/1 (i.e., reject/accept) decision. Always accepting an order when capacity is available could unable the system to accept more convenient orders in the future. Another important aspect is the aV'(tiiability of information to the decisionmaker. We use a stochastic modeling approach using Markov decision theory and learning methods from Artificial Intelligence techniques in order to deal with uncertainty and long-term decisions in Ok Reinforcement Learning (RL) is a quite new approach that already combines this idea of modeling and solution method. Here we report on RL-solutions for some OA models.

AB - Order Acceptance (OA) is one of the main functions in a business control framework. Basically, OA involves for each order a 0/1 (i.e., reject/accept) decision. Always accepting an order when capacity is available could unable the system to accept more convenient orders in the future. Another important aspect is the aV'(tiiability of information to the decisionmaker. We use a stochastic modeling approach using Markov decision theory and learning methods from Artificial Intelligence techniques in order to deal with uncertainty and long-term decisions in Ok Reinforcement Learning (RL) is a quite new approach that already combines this idea of modeling and solution method. Here we report on RL-solutions for some OA models.

KW - IR-95693

M3 - Report

T3 - BETA working paper

BT - Order acceptance with reinforcement learning

PB - University of Twente, Research School for Operations Management and Logistics (BETA)

CY - Enschede, the Netherlands

ER -

Mainegra Hing M, van Harten A, Schuur P. Order acceptance with reinforcement learning. Enschede, the Netherlands: University of Twente, Research School for Operations Management and Logistics (BETA), 2001. 44 p. (BETA working paper; WP-66).