Estimating manufacturing electricity costs by simulating dependence between production parameters

Hyun Woo Jeon, Seok Gi Lee, Chao Wang

Research output: Contribution to journalArticle

Abstract

The electricity costs associated with manufacturing generally consist of an energy charge and a demand charge, which are calculated by total kWh × $/kWh and peak kW × $/kW, respectively. While these two charges have been extensively studied with production scheduling, dependence between production parameters has not been thoroughly investigated for production energy costs when the production parameters are assumed to be independent. Such dependence, however, can significantly affect the peak kW and resulting energy costs. Therefore, industrial practitioners need a tool to simulate the production power demand and cost to evaluate energy costs for each production plan and for any dependence in the parameters. In this paper, we propose a method for simulating the power demand at the machine and facility levels, focusing on the dependence in milling machine parameters. Using probabilistic techniques with copulas, we quantify the degree of dependence between two milling parameters (processing times and processing amounts) and build a power demand simulator combining discrete event simulation and numerical simulation models. From 35 illustrative examples, we show that a production strategy to adjust dependence can reduce the peak kW and energy costs by more than 15% and 4%, respectively. In addition to dependence, less variability in processing times results in less peak kW and energy costs. Other factors such as a marginal probability distribution and copula types having the same degree of variability have been found to have limited effects on reducing energy costs.

LanguageEnglish (US)
Pages129-140
Number of pages12
JournalRobotics and Computer-Integrated Manufacturing
Volume55
DOIs
StatePublished - Feb 1 2019

Fingerprint

Electricity
Manufacturing
Costs
Energy
Charge
Copula
Processing
Production/scheduling
Milling machines
Discrete event simulation
Discrete Event Simulation
Marginal Distribution
Probability distributions
Simulation Model
Simulator
Quantify
Probability Distribution
Simulators
Scheduling
Numerical Simulation

Keywords

  • Correlation
  • Dependence
  • Electricity costs
  • Milling power
  • Peak demand

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Mathematics(all)
  • Computer Science Applications
  • Industrial and Manufacturing Engineering

Cite this

Estimating manufacturing electricity costs by simulating dependence between production parameters. / Jeon, Hyun Woo; Lee, Seok Gi; Wang, Chao.

In: Robotics and Computer-Integrated Manufacturing, Vol. 55, 01.02.2019, p. 129-140.

Research output: Contribution to journalArticle

@article{205711f4d12e49f09ad034cafc5343ae,
title = "Estimating manufacturing electricity costs by simulating dependence between production parameters",
abstract = "The electricity costs associated with manufacturing generally consist of an energy charge and a demand charge, which are calculated by total kWh × $/kWh and peak kW × $/kW, respectively. While these two charges have been extensively studied with production scheduling, dependence between production parameters has not been thoroughly investigated for production energy costs when the production parameters are assumed to be independent. Such dependence, however, can significantly affect the peak kW and resulting energy costs. Therefore, industrial practitioners need a tool to simulate the production power demand and cost to evaluate energy costs for each production plan and for any dependence in the parameters. In this paper, we propose a method for simulating the power demand at the machine and facility levels, focusing on the dependence in milling machine parameters. Using probabilistic techniques with copulas, we quantify the degree of dependence between two milling parameters (processing times and processing amounts) and build a power demand simulator combining discrete event simulation and numerical simulation models. From 35 illustrative examples, we show that a production strategy to adjust dependence can reduce the peak kW and energy costs by more than 15{\%} and 4{\%}, respectively. In addition to dependence, less variability in processing times results in less peak kW and energy costs. Other factors such as a marginal probability distribution and copula types having the same degree of variability have been found to have limited effects on reducing energy costs.",
keywords = "Correlation, Dependence, Electricity costs, Milling power, Peak demand",
author = "Jeon, {Hyun Woo} and Lee, {Seok Gi} and Chao Wang",
year = "2019",
month = "2",
day = "1",
doi = "10.1016/j.rcim.2018.07.009",
language = "English (US)",
volume = "55",
pages = "129--140",
journal = "Robotics and Computer-Integrated Manufacturing",
issn = "0736-5845",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Estimating manufacturing electricity costs by simulating dependence between production parameters

AU - Jeon, Hyun Woo

AU - Lee, Seok Gi

AU - Wang, Chao

PY - 2019/2/1

Y1 - 2019/2/1

N2 - The electricity costs associated with manufacturing generally consist of an energy charge and a demand charge, which are calculated by total kWh × $/kWh and peak kW × $/kW, respectively. While these two charges have been extensively studied with production scheduling, dependence between production parameters has not been thoroughly investigated for production energy costs when the production parameters are assumed to be independent. Such dependence, however, can significantly affect the peak kW and resulting energy costs. Therefore, industrial practitioners need a tool to simulate the production power demand and cost to evaluate energy costs for each production plan and for any dependence in the parameters. In this paper, we propose a method for simulating the power demand at the machine and facility levels, focusing on the dependence in milling machine parameters. Using probabilistic techniques with copulas, we quantify the degree of dependence between two milling parameters (processing times and processing amounts) and build a power demand simulator combining discrete event simulation and numerical simulation models. From 35 illustrative examples, we show that a production strategy to adjust dependence can reduce the peak kW and energy costs by more than 15% and 4%, respectively. In addition to dependence, less variability in processing times results in less peak kW and energy costs. Other factors such as a marginal probability distribution and copula types having the same degree of variability have been found to have limited effects on reducing energy costs.

AB - The electricity costs associated with manufacturing generally consist of an energy charge and a demand charge, which are calculated by total kWh × $/kWh and peak kW × $/kW, respectively. While these two charges have been extensively studied with production scheduling, dependence between production parameters has not been thoroughly investigated for production energy costs when the production parameters are assumed to be independent. Such dependence, however, can significantly affect the peak kW and resulting energy costs. Therefore, industrial practitioners need a tool to simulate the production power demand and cost to evaluate energy costs for each production plan and for any dependence in the parameters. In this paper, we propose a method for simulating the power demand at the machine and facility levels, focusing on the dependence in milling machine parameters. Using probabilistic techniques with copulas, we quantify the degree of dependence between two milling parameters (processing times and processing amounts) and build a power demand simulator combining discrete event simulation and numerical simulation models. From 35 illustrative examples, we show that a production strategy to adjust dependence can reduce the peak kW and energy costs by more than 15% and 4%, respectively. In addition to dependence, less variability in processing times results in less peak kW and energy costs. Other factors such as a marginal probability distribution and copula types having the same degree of variability have been found to have limited effects on reducing energy costs.

KW - Correlation

KW - Dependence

KW - Electricity costs

KW - Milling power

KW - Peak demand

UR - http://www.scopus.com/inward/record.url?scp=85051638296&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051638296&partnerID=8YFLogxK

U2 - 10.1016/j.rcim.2018.07.009

DO - 10.1016/j.rcim.2018.07.009

M3 - Article

VL - 55

SP - 129

EP - 140

JO - Robotics and Computer-Integrated Manufacturing

T2 - Robotics and Computer-Integrated Manufacturing

JF - Robotics and Computer-Integrated Manufacturing

SN - 0736-5845

ER -