4 Presentations

• 
  10:30 AM - 11:00 AM

  Optimal Pricing of Retail Gasoline

  • Daero Kim, presenter, The University of Western Ontario
  • Matt Davison, Departments of Applied Mathematics and Statistical & Actuarial Sciences, Richard Ivey School of Business, The University of Western Ontario
  • Fredrik Odegaard, The Ivey School of Business, The University of Western Ontario

  We present models for analyzing demand of retail gasoline and derive optimal pricing strategies in an oligopoly setting. In order to study the behaviour of frequently changing gasoline price in competitive market, one need to collect a data from a large geographic area that contains one relevant competitive market for retail gasoline stations. We obtained the station specific daily data on gasoline prices and demand for 100 different sites each with 6 to 8 competitors in cities for a year. We will empirically test each model and compare the results. We expect the piecewise quadratic demand function will produce more meaningful results than the linear demand function. In the application of the models, we will also attempt to analytically find retailer’s real-time optimal strategy and compute the forecasted price, demand and profit compared with the actual data, using autoregressive integrated moving average (ARIMA) models with Kalman filtering and smoothing.

• 
  11:00 AM - 11:30 AM

  Investigation of Energy Consumption Indicator of Buildings by Clustering and ANOVA Analysis

  • Wen-Shing Lee, presenter, National Taipei University of Technology
  • Lung-Chieh Lin, National Taipei University of Technology

  Making energy consumption indicator for buildings is important for energy management. The general indicator of building energy consumption is the energy usage per unit area, EUI, kWh/m2/year. Many Factors, climate, building type, using type, number of staff and customers, affect the value of EUI for buildings. This paper uses ANOVA analysis to distinguish the buildings to different groups. First, the evaluated buildings are cluster to different group by using type, such as Township Office, Household Registration Office, Land Office. Then, the regression coefficient of staff factor in a liner regression, using energy consumption as dependent variable and number of staff as Independent variable, of different group and different climate of buildings is calculated. Finally, a two way ANOVA test is used to determine whether the clustering is valid or not. The result can use to clustering the buildings and energy consumption indicators can make for different grouping buildings. A case study in Taiwan is analyzed in this paper.

• 
  11:30 AM - 12:00 PM

  Risk Hedging Optimal Capacity in the European Hub-Based Natural Gas Market Network: A Model-Based Approach

  • Parviz Darvish, presenter, ESSEC Business School

  Hub-based natural gas (NG) trades are growing in importance in Europe and may attain a considerable share of the total European NG trade in the future. Growing trading in European hubs, rather than in the form of long-term contracts, makes the market more liquid and competitive, while causing the market to become more volatile and susceptible to risk. In a hub-based market, namely, the futures and spot trade framework, the production and transportation capacities needed to fulfill the real final demand and achieve the equilibrium are quite uncertain. These capacities are functions of other decision variables, such as equilibrium flows and prices in the presence of stochasticity throughout the market network. Throughout this paper, we have relied on five sub-models comprising market structure, nonlinear optimization, game theory, Conditional Value at Risk (CVaR), and simulation. We conducted nonlinear optimization programming for a restructured NG market network, given the demand side uncertainty of the final consumers in spot trade. We consider a two-stage market, corresponding to futures and spot trades, in which we make decisions in the futures market subject to uncertainty in the spot market. The nonlinear programming was developed relying on the Mixed Complementary Problem (MCP), in terms of the simultaneous presence of numerous players in this market network, such as producers, exporters, wholesale traders, network operators, and consumers, and the model also allows for the game theoretic behaviour on the part of the participants. The proposed algorithm incorporates a simulation-based iterated procedure for achieving well-converged results. We ran the model for a synthetic NG network, resembling the European market network, which comprises three major suppliers, three main wholesale traders, and three consumers and corresponding network operators. The results for optimal risk hedging capacities for pipelines compared with current capacities represent abundance or shortage in each corresponding link in the network. These capacity shortages serve as signals for inventory or new investments. We applied the same procedure to production capacities to achieve the optimal ex ante risk hedging decision at the production points. The model also suggests quantitative resolutions for the confidence level of the random decision variables versus demand uncertainty, such that increasing both the uncertainty and confidence level of the decision variable leads to allocating added capacities against risk hedging.

• 
  12:00 PM - 12:30 PM

  Using Real Option Analysis to Quantify Ethanol Policy Impact on the Firm's Entry Into and Optimal Operation of Corn Ethanol Facilities

  • Christian Maxwell, presenter, University of Western Ontario

  Ethanol crush spreads are used to model the value of a facility which produces ethanol from corn. A real options analysis is used to investigate the effects of model parameters on management's decision to operate the facility through optimal switching and the firm's decision to enter into the project given its expected real options net present value. We perform the analysis by means of a stochastic optimal control problem. We present evidence of increased correlation between corn and ethanol prices, perhaps as a result of government policy which has induced more players to enter into the market. This talk investigates the subsequent negative effects on firms. Further, this talk illustrates the impact of an abrupt change in government policy, as happened in January 2012, on a firm's decision to enter the business.