ICAOR’11 ABSTRACTS

3RD INTERNATIONAL CONFERENCE ON APPLIED OPERATIONAL RESEARCH

24-26 AUGUST 2011, ISTANBUL, TURKEY


OPTIMIZATION OF NURSE ROSTERS IN A HOSPITAL UNIT

Andreia Filipa Antunes

Centro de Investigação Operacional, Universidade de Lisboa, Lisboa, Portugal

Margarida Moz

Universidade Técnica de Lisboa and Centro de Investigação Operacional, Universidade de Lisboa, Lisboa, Portugal

Abstract. The present work reports on the application of a mixed integer linear programming model to nurse rostering in the real context of a hospital unit in Portugal. The comparison of the results obtained with those obtained with the current procedure in the unit shows that this is a promising approach to tackle the problem in this context.


A MULTI-CLASS OPEN QUEUING NETWORK WITH A PRIORITY DISCIPLINE: A CASE STUDY OF A LOCAL EMERGENCY CARE CENTRE

Sumi Kim and Seongmoon Kim

School of Business, Yonsei University, Seoul, Korea

Abstract. Unlike ordinary outpatient clinics, an emergency care centre has non-homogeneous patients with diverse diseases of different acuity levels. Since patients in critical condition face serious consequences, including death, the target waiting times should be determined based on their acuity levels. To reflect the special situation of the emergency care centre, we formulate patient flows using a multi-class open queuing network. This paper is unique because of its integration of a priority discipline with a queuing network to control the waiting times for more urgent patients. A case study based on actual data from an emergency care centre demonstrates that the priority policy is effective in decreasing the waiting times for the higher-priority classes without significantly sacrificing service qualities for the lower-priority classes.


DE.LO.S.: A DECISION SUPPORT SYSTEM FOR SOLVING SUPPLY CHAIN MANAGEMENT PROBLEMS

Yannis Marinakis and Magdalene Marinaki

Technical University of Crete, Crete, Greece

Abstract. In this paper a Decision Support System (DSS), called De.Lo.S. (Decisions in Logistics and Supply Chain Management) is introduced, which is designed for assisting in the integral control of the logistic activities in a firm. De.Lo.S. has the possibility of solving either a logistic problem, such as Vehicle Routing Problem, Inventory Control Problem, Location Problem, etc., separately, or to combine some or all of these problems for an integrated solution. De.Lo.S. consists firstly of a Graphical User Interface that allows the effective interaction between the system and the user, secondly of a data base that stores all the necessary information needed for the DSS, and thirdly of a model base that contains the models and the algorithms needed in order to provide the suitable solution.


OPTIMIZATION MODELS IN THE LOCATION OF HEALTHCARE FACILITIES: A REAL CASE IN BRAZIL

Arnaldo Rabello de Aguiar Vallim Fo. and Iara da Silva Mota

College of Computing and Informatics, Mackenzie Presbyterian University, São Paulo, Brasil

Abstract. This paper addresses the question of locating healthcare facilities, with the aid of some of the most important location models: P-median model, set and maximal covering models and P-centre model. An experiment was carried out in a real situation, in a Brazilian city, seeking to confront the current sites configuration with optimal solutions proposed by the location models. Valuable results have been generated, showing the effectiveness of the models and permitting to highlight the strong points of each approach. The paper presents a discussion of all these results and describes the case under study, the models and the experiment.


EFFICIENCY EVALUATION OF THE INDIAN BUS COMPANIES USING DEA IN THE LIGHT OF WEIGHT RESTRICTIONS

Punita Saxena

University of Delhi, Delhi, India

Abstract. This paper examines the performance of the Indian bus companies using Data Envelopment Analysis. This non-parametric approach of efficiency evaluation is based on piecewise linear frontiers estimated with the help of mathematical programming techniques. This technique is used to evaluate the technical, pure technical and managerial efficiencies of the Indian bus companies. Data was obtained from the publications of Central Institute of Road Transport (CIRT) of India. Since weight flexibility is a major drawback in using the CCR and BCC models of DEA, hence, weight restrictions have also been introduced in the DEA models. The analysis has shown bus companies with clear differences in their productivity.


INTEGER LINEAR PROGRAMMING MODELS FOR MEDIA STREAMS PLANNING

Pavel Troubil and Hana Rudová

Faculty of Informatics, Masaryk University, Brno, Czech Republic

Abstract. Advanced collaborative environments frequently need to transfer highly demanding multimedia data streams with minimum possible latency. Since bandwidths of the streams are close to capacities of currently available links, routing of data transfers therefore requires planning with respect to link capacities and latency optimization. This paper describes integer linear programming techniques for optimal solution of the multimedia streams planning problem. Several methods for cycle avoidance in transmission graphs and network flows formulation are proposed. Performance of the methods was evaluated to identify their efficiency for real-time planning and to compare it against an earlier constraint programming approach applied in an application middleware called CoUniverse. According to the results, network flows formulation appears to be the most promising.


SECURITY PERSONNEL ROUTING AND ROSTERING: A HYPER-HEURISTIC APPROACH

Mustafa Misir, Pieter Smet, Katja Verbeeck and Greet Vanden Berghe

KAHO Sint-Lieven, Gent, Belgium

Katholieke Universiteit Leuven Campus Kortrijk, Kortrijk, Belgium

Abstract. In the present study, a large scale, structured problem regarding the routing and rostering of security personnel is investigated. Structured problems are combinatorial optimization problems that encompass characteristics of more than one known problem in operational research. The problem deals with assigning the available personnel to visits associated with a set of customers. This objective just described, reflects the rostering characteristic of the problem. In addition, the different geographic locations of the customers indicate the requirement of routing. A new benchmark dataset for this complex problem is presented. A group of high-level problem-independent methods, i.e. hyper-heuristics, is used to solve this novel problem. The performance and behaviour of different hyper-heuristics for the presented benchmark dataset are analysed.


INVESTMENT PERFORMANCE OF THE MARKOWITZ‟S PORTFOLIO SELECTION MODEL IN THE KOREAN STOCK MARKET

Seongmoon Kim and Hongseon Kim

School of Business, Yonsei University, Seoul, Korea

Abstract. This paper investigated performance of the Markowitz‟s portfolio selection model with applications to Korean stock market. We chose Samsung-Group-Funds and KOSPI (Korea Composite Stock Price Index) for performance comparison with the Markowitz‟s portfolio selection model. For one and a half year period between March 2007 and September 2008, KOSPI almost remained the same with only 0.1% change, Samsung-Group-Funds showed 20.54% return, and Markowitz‟s model, which is composed of the same 17 Samsung group stocks, achieved 52% return. We did sensitivity analysis on the length of referencing financial database (6 months, 1 year, and 2 years) and the frequency of portfolio change (1 week, 4 weeks, 8 weeks, and 12 weeks) to compare the investment performance.


COMPARISON OF PROJECT MANAGEMENT SOFTWARE PACKAGES FOR RESOURCE CONSTRAINED PROJECT SCHEDULING PROBLEM

Yavuz Gunalay

Bahcesehir University, Istanbul, Turkey

Erdem Levent

Eregli Demir Celik Inds, Istanbul, Turkey

Abstract. In this study, it is aimed to compare the performance of resource levelling tools of project management software packages on the solution of resource constrained project scheduling. Two software packages, one commercial, MS Project, and one open source, Open Workbench, are investigated by using two resource constrained R&D projects. Results are also compared with a previous study published in International Journal of Project Management. In conclusion, it is seen that performance of resource levelling tools of software packages are independent from the characteristics of projects investigated.


A MIXED INTEGER PROGRAMMING APPROACH FOR DATA CLASSIFICATION BASED ON CASE-BASED DISTANCE METHOD

Li-Ching Ma

National United University, MiaoLi, Taiwan

Abstract. Classification is a procedure to separate data or alternatives into two or more than two classes. The need to classify alternatives involving multiple criteria into one of distinct classes is considerable in practice. Therefore, determining how to assist decision makers in classification is an important issue of multi-criteria decision aids. This study tries to develop a classification model incorporating the advantages of the cases-based distance method and mixed integer programming model. The proposed classification model can classify alternatives by evaluating sets of cases selected by decision makers, reduce misclassification rate, and improve impact of outliers. The proposed classification approach is demonstrated using an inventory planning and control example.


AN EVOLUTIONARY SIMULATION-OPTIMIZATION APPROACH FOR ADDRESSING UNMODELLED OBJECTIVES IN WASTE MANAGEMENT FACILITY EXPANSION PLANNING

Julian Scott Yeomans

York University, Toronto, ON Canada

Yavuz Gunalay

Bahcesehir University, Istanbul, Turkey

Abstract. Public sector decision-making typically involves complex problems that are riddled with competing performance objectives and possess design requirements which are difficult to capture at the time that supporting decision models are constructed. Environmental policy formulation can prove additionally complicated because the various system components often contain considerable stochastic uncertainty and frequently there are also numerous stakeholders holding incompatible perspectives. Consequently, there are invariably unmodelled performance design issues, not apparent at the time of the problem formulation, which can greatly impact the acceptability of any proposed solutions. While a mathematically optimal solution might provide the best solution to a modelled problem, normally this will not be the best solution to the underlying real problem. Therefore, in public environmental policy formulation, it is generally preferable to be able to create several quantifiably good alternatives that provide very different approaches and perspectives to the problem. This study shows how simulation-optimization (SO) modelling can be combined with a niching approach to efficiently generate multiple policy alternatives that satisfy required system performance criteria in stochastically uncertain environments and yet are maximally different in the decision space. The efficacy of this modelling-to-generate-alternatives method is specifically demonstrated on a municipal solid waste management facility expansion case.


MULTI-OBJECTIVE PORTFOLIO SELECTION PROBLEMS IN A SOFT COMPUTING FRAMEWORK

José Vicente Segura

Universidad Miguel Hernández de Elche, Alicante, Spain

José D. Bermúdez

Universitat de València, Valencia, Spain

Enriqueta Vercher

Universitat de València, Valencia, Spain

Abstract. In this paper we present a multi-objective decision approach for the fuzzy portfolio selection problem, assuming that uncertainty is modeled by means of fuzzy logic. Specifically, we approach the uncertainty on the future returns of the investment using LR-fuzzy numbers and consider some f-weighted interval-valued possibilistic moments to approximate the return and risk of a given portfolio. A genetic procedure finds the efficient frontier of certain possibilistic mean-downside risk-skewness models. A numerical experiment is reported for assets from the Spanish stock market.


PATIENT SCHEDULING IN RADIOTHERAPY PLANNING

S Sheibani

Faculty of Management, University of Tehran, Iran

K Sheibani

Tadbir Operational Research Group, Vancouver, BC Canada

Abstract. This paper describes a very fast heuristic for a real-life scheduling problem arising at a major local radiotherapy clinic. After referral to radiotherapy, patients need to visit a set of certain facilities sequentially within a specific due date before starting a treatment scheme. Each patient requires each facility at most once for a roughly determined amount of time. These operations are traditionally recognised as the radiotherapy planning. The main objective is to minimise the mean flow time of the patients in the radiotherapy planning. The problem can be seen as a dynamic job shop environment. The proposed method consists of two phases: building a list of patients and constructing a schedule. A fuzzy greedy evaluation function is employed to prioritise the operations in the list for incorporating into the construction phase of the heuristic. Computational experiments on standard benchmark problems indicate that the proposed method is very effective. Monte Carlo simulation will be also applied to validate the proposed scheduling model.


WELL-POSEDNESS OF PORTFOLIO CHOICE PROBLEMS WITH RISK MEASURES AND FRICTIONS

Alejandro Balbás

University Carlos III of Madrid, Madrid, Spain

Beatriz Balbás

University of Castilla La Mancha, Toledo, Spain

Raquel Balbás

University Complutense of Madrid. Madrid, Spain

Abstract. This paper studies a portfolio choice problem such that the pricing rule may incorporate transaction costs and the risk measure is coherent and expectation bounded. We will prove the necessity of dealing with pricing rules such that there exists an essentially bounded stochastic discount factor, which must be also bounded from below by a strictly positive value. Otherwise good deals will be available to traders, i.e., depending on the selected risk measure, investors can build portfolios whose (risk, return) will be as close as desired to (-infinity, infinity) or (0, infinity). This pathologic property still holds for vector risk measures (i.e., if we minimize a vector valued function whose components are risk measures). It is worthwhile to point out that essentially bounded stochastic discount factors are not usual in financial literature. In particular, the most famous frictionless, complete and arbitrage free pricing models imply the existence of good deals for every coherent and expectation bounded (scalar or vector) measure of risk, and the incorporation of transaction costs will not guarantee the solution of this caveat.


NEURAL NETWORK BASED IDENTIFICATION FOR COMPLEX NON-LINEAR SYSTEMS

Otu Vaarmann

Institute of Cybernetics, Tallinn University of Technology, Tallinn, Estonia

Abstract. This report treats numerical methods for highly non-linear least squares problems for which procedural and rounding error are unavoidable, e.g. those arising in the development of various non-linear system identification techniques based on input-output representation of the model such as training of artificial neural networks. The computational aspects are discussed and a local convergence theorem is proposed.


A MARKOV DECISION MODEL TO OPTIMIZE HOSPITAL BED CAPACITY UNDER STOCHASTIC DEMAND

Paul Kizito Mubiru

Kyambogo University, Kyambogo, Uganda

Abstract. Hospitals continually face the challenge of planning and managing bed capacities within an environment of demand uncertainty. In this paper, an optimization method for allocating hospital bed capacities is proposed. The model, based on Markov decision process approach, matches demand to bed availability levels of a health care system. Adopting such an approach, the states of a Markov chain represent possible states of demand for bed occupancy. The decision of whether or not to admit additional patients is made using dynamic programming .This approach demonstrates the existence of an optimal state-dependent capacity level, and produces an optimal admission policy for patients as well as the corresponding total capacity costs.


QUANTIFYING THE BENEFITS OF PHASING AS A CORPORATE REAL ESTATE STRATEGY

Mikael Collan

University of Turku, School of Economics, Turku, Finland

Abstract. The purpose of this paper is to show how it is possible to quantify the benefits of phasing in corporate real estate projects in a way that makes understanding project risks intuitive and enhances the support for project decision-making. When the potential and the risk in different real estate strategies is clearly visible and quantifiable, the potentially significant risk management aspect of phasing becomes easier to appreciate and to quantify. Quantification of benefits from phasing corporate real estate projects can be very difficult without the proper tools, this case study shows for the first time how a quantitative analysis on a corporate real estate project can be performed simply with the use of cash-flow scenario based pay-off distributions. The method presented is a good addition to the toolkit of the real estate industry and corporate real estate managers.


DID THE LEHMAN SHOCK HAVE AN IMPACT ON CONSISTENCY OF RATING INFORMATION?

Motohiro Hagiwara

Meiji University, Tokyo, Japan

Abstract. The purpose of this research is to test the informational consistency of long-term bond rating information and sovereign ratings information by main four rating agencies in Japan and America. This study attempts to rate the ratings and compare rating information by each agency from the perspective of informational consistency. Informational time consistency is an important characteristic that an ideal rating information should fulfil. Consistency means that, if factors that determines the rating information are the same, each rating agency publish the same rating in different time and sectors. The most important point of analysis is to test the time consistency of rating before and after the Lehman Shock and verify quantitatively the validity of appraisal and doubts on the method of determining the rating that occurred after the Lehman Shock. As most important finding, the downgrading during the period that experienced increased international credit risks after the Lehman Shock in 2008, is due to changes in the fundamental, and consistency of the rating character was maintained. However, only in the case of corporate ratings by JCR was there a downward bias in rating for the same characteristic values after Lehman Shock.


ELECTRICITY DEMAND FORECASTS USING GENERALIZED EXPONENTIAL SMOOTHING MODELS

José D. Bermúdez

Universitat de València, Valencia, Spain

Ana Corberán-Vallet

Medical University of South Carolina, Charleston, South Carolina USA

José V. Segura

Universidad Miguel Hernández de Elche, Elche, Spain

Enriqueta Vercher

Universitat de València, Valencia, Spain

Abstract. We introduce an extension of exponential smoothing to deal with covariates and double seasonality that could easily be adapted to more than two seasonal cycles. Assuming additive effects and a stochastic component given by independent, homoscedastic normal errors, the exponential smoothing model can be expressed as an equivalent linear dynamic model with a very peculiar structure of the covariance matrix. The covariance matrix is a function of the unknown smoothing parameters only, while the mean vector only depends on the unknown initial conditions. These facts allow for a simplification of the statistical analysis of the model. Following the Bayesian paradigm, we obtain the joint posterior distribution of all the unknowns. Only the marginal posterior of the smoothing parameters is analytically intractable and has to be approached using simulation techniques. The conditional distribution of initial conditions giving the smoothing parameters is well-known and it can be integrated out exactly in order to compute the predictive distribution. Finally, we propose to integrate out the smoothing parameters using Monte-Carlo techniques, obtaining an estimate of the predictive distribution as well as their main characteristics: point forecasts and prediction intervals. We apply this methodology to electricity demand forecasts using two real data sets.


ON THE TRANSPORTATION FLEET MAINTENANCE SCHEDULING PROBLEM OF A MULTI-FACILITY MAINTENANCE SYSTEM WITH LIMITED CAPACITY

Jia-Yen Huang

National Chin-Yi University of Technology, Taichung, Taiwan

Ming-Jong Yao

National Chiao Tung University, Hsinchu, Taiwan

Abstract. In the real world, the managers of a transportation fleet need to carefully determine the maintenance cycles of vehicles so as to minimize the average total costs (including the operating costs and the maintenance costs). Furthermore, since the maintenance system usually equips with limited manpower and facilities, it has limited capacity to provide maintenance service for the transportation fleet. In this paper, a new and efficient heuristic to solve the transport fleet maintenance scheduling problem in a multi-facility maintenance process with limited capacity is presented. To solve this problem, under the assumption that the maintenance cycle times are integer multiples of a basic period, a search algorithm to obtain the relaxed optimal solution for an unconstrained model is developed. Once a candidate solution of the time multiples are chosen, we determine for each basic period of the global cycle the group of vehicles to be maintained and the sequence to be used. Then a feasibility testing heuristic is performed for each one of these candidate solutions. Finally, among these candidate solutions, we pick the „best‟ solution that secures a feasible maintenance schedule with the minimum average costs. Based on our experiments, we demonstrate that the proposed search algorithm could effectively solve the Transportation Fleet Maintenance Scheduling Problem in a multi- facility maintenance system with limited capacity.


A MATHEMATICAL MODEL FOR AMBULANCE LOCATION AND DEPLOYMENT

F Samanlioglu, Z Ayag, A Yucekaya, A Kilic, I Balkuvar and Z Cikrikci

Kadir Has University, Istanbul, Turkey

Abstract. Response time is crucial in case of events that require emergency medical services (EMS) such as illnesses, large-scale emergency needs, accidents, etc. since this service is directly related with life threatening incidences. Response time is mostly related to the location and deployment of EMS. In this paper, a mathematical model is developed to locate and deploy ambulance stations so that all demand areas are covered in terms of “primary” or “secondary” coverage. These terms are defined based on the response time of ambulances in case of an event. In the model, two types of ambulance stations are considered, independent stations located outside health facilities, and stations located inside health facilities. The mathematical model is applied in selected areas of Istanbul and solved optimally with LINGO 7.0 solver.


OPTIMAL PRODUCTION MANAGEMENT WHEN DEMAND DEPENDS ON BUSINESS CYCLES

Abel Cadenillas

University of Alberta, Edmonton, Canada

Peter Lakner and Michael Pinedo

Stern School of Business, New York University, New York USA

Abstract. In this paper we assume that consumer demand for an item follows a Brownian motion with a drift that is modulated by a continuous-time Markov chain that represents the regime of the economy. The economy may be in either one of two regimes; one regime may represent a recession period with a low demand rate and the other may represent an expansion period with a high demand rate. The economy remains in one regime for a random amount of time that is exponentially distributed with a given rate and then moves to the other regime and remains there for an exponentially distributed amount of time with another rate. The management of the company would like to maintain the inventory level of the item as close as possible to a target inventory level and would also like to produce the items at a rate that is as close as possible to a target production rate. The company is penalized by the deviations from the target levels and the objective is to minimize the long term total expected discounted penalty costs. We consider two models. In the first model the management of the company observes the regime of the economy at all times (case of full information), whereas in the second model the regime of the economy is unobserved (case of limited information). We solve both problems by applying the technique of “completing squares” and obtain formulas for the optimal production policy as well as for the minimal total expected discounted cost. Our analytical results show, among various other results, that in both models the optimal production policy depends on factors that are based on short term concerns as well as factors that are based on long term concerns. We analyze how the impact of these factors depend on the values of the parameters in the model. We furthermore compare the total expected discounted costs of the two models and determine the value of information concerning the current regime of the economy. Among other results we show that the difference of the costs in the two models is proportional to the total discounted expected squared error of estimate of the regime. We show that under a quadratic utility in both the full and the limited information cases the optimal production rate is a linear function of the current inventory level. In the full information case the coefficients in this linear function depend on the current state of the economy, and we present an explicit representation for the optimal production rate and the minimal level of penalty cost. In the limited information case the coefficients are the conditional expectations of the corresponding coefficients in the full information case, given the available information at the time. We compute a more explicit formula than just this conditional expectation, using filtering theory applied to Hidden Markov Models. In this case our solution is not as explicit as in the full information case as it is a function of the solution of a two-dimensional linear stochastic differential equation (SDE). Numerical solution of SDE's is not a trivial matter. However, we can replace these SDE's by a system of linear homogeneous ordinary differential equations (ODE) of the first order, and the numerical solution of such system of ODE's is standard.


PRICE TAKER BIDDING FOR POWER PLANTS IN TURKISH POWER MARKET

Ahmet Yücekaya, Zeki Ayag, Funda Samanlioglu

Kadir Has University, Istanbul, Turkey

Abstract. In Turkish power market, power firms bid daily into the day-ahead power market. Forecasted demand for each hour of the next day is announced so as the power supplier can submit offer for the demand. The auction mechanism and competition determine the equilibrium price and quantity for each hour. If the bid price of a company is below the market price, the offer of the company will be accepted and will be rewarded with the market price. The profit maximization for price-taker units under price uncertainty and blind auction rules requires careful preparations of bidding offers. In this paper, we use a normal distribution based market price generation method and a marginal cost based bid generation method to simulate the market for the company. The bidding strategy that will maximize the expected profit over price scenarios is selected to be submitted to the market. We illustrate the model for a coal unit in Turkish power market and present the results.


AN IMPROVEMENT IN THE STOCHASTIC STOCK PRICE DIFFERENCES MODEL

Susumu Saito and Miyuki Hasegawa

School of Management, Tokyo University of Science, Tokyo, Japan

Abstract. An improvement of the equation in a stochastic model for stock price and foreign exchange simulations was tried. In the improved equation, the random walk term in the stochastic model is multiplied by the absolute value of the difference between the current price and the price in the previous period divided by square root of a time interval raised to the power of a certain value, plus a term which is the price difference multiplied by a coefficient. As a result, a differences distribution close to an actual distribution with a high peak and fat tails was obtained, and a financial time series simulation model whose differences distribution, option prices and final prices remain the same irrelevant of the length of time interval was derived. The derived time series also show movements close to actual movements in that they have continuous small or large fluctuations and continuous price increase or decrease.


OPTIMIZING THE ROUTE OF THE ARM OF AN ASSEMBLY ROBOT

Hajieh Jabbari K. and Béla Vizvári

Eastern Mediterranean University, Mersin, Turkey

Abstract. Productivity of an assembly robot depends on the route length of its arm. One approach to minimize the move is to model the problem as a Travelling Salesman Problem (TSP) in a bipartite graph. One part of the vertices consists of the cells containing the components to be assembled and the other part is the set of assembly positions. Production of Printed Circuit Board (PCB) is a typical example of such a problem. In this case, the robot practically moves in a plane but the resulted TSP differs from the TSPs on the plane having connection between any two vertices. In this study fast heuristics, branch and cut solution as constructive heuristics, and special mathematical properties of the problem are discussed.


CONVEX EXTENSION OF DISCRETE-CONVEX FUNCTIONS AND APPLICATIONS IN OPTIMIZATION THE STRUCTURE OF PARALLEL AND DISTRIBUTED PROCESSING SYSTEMS

Tiit Riismaa

Institute of Cybernetics, Tallinn University of Technology, Tallinn, Estonia

Abstract. A method of description and optimization of the structure of multi-level parallel and distributed processing systems is presented. The set of feasible structures for such class of systems is defined. The representation of this set is constructed in terms of the graph theory. A recursive algorithm is constructed to solve the general problem of optimal multi-level paralleling procedure. For the reduced statement two types of variable parameters are defined: for the level size and for the relations of adjacent levels. Two different classes of iteration methods are developed. For solving the reduced problem the recursive algorithm is constructed, where index of level is the index of recursion. Also a numerical method of local searching is developed. On each step of the iteration the calculation of the value of objective function is required only on some vertices of some kind of unit cube. Modelling and optimization of the structure of multi-level processing system illustrate the considered approach.


A FUZZY PAY-OFF METHOD FOR REAL OPTION VALUATION: CREDIBILISTIC APPROACH

Mikael Collan, Robert Fuller and Jozsef Mezei

University of Turku, Turku School of Economics, Turku, Finland,

IAMSR, Abo Akademi University, Turku, Finland

Turku Centre for Computer Science, Turku, Finland

Abstract. Real option analysis offers interesting insights on the value of assets and on the profitability of investments, which has made real options a growing field of academic research and practical application. Real option valuation is, however, often found to be difficult to understand and to implement due to the quite complex mathematics involved. Recent advances in modelling and analysis methods have made real option valuation easier to understand and to implement. This paper extends the results of our earlier paper on fuzzy pay-off method for real option valuation by using credibility measures.


SOLUTION PROCEDURES FOR THE RECTILINEAR DISTANCE SINGLE SOURCE CAPACITATED MULTI-FACILITY WEBER PROBLEM

Temel Öncan, M. Hakan Akyüz, I. Kuban Altinel and M. Emre Demircioglu

Galatasaray University, Istanbul, Turkey

Bogaziçi University, Istanbul, Turkey

Abstract. In this work we consider the Single Source Capacitated Multi-facility Weber Problem (SSCMWP) which deals with locating I capacitated facilities in the plane in order to satisfy the demands of J customers for a single commodity type so that each customer satisfies all its demand from a single facility and the total transportation cost is minimized. We address the rectilinear case of the SSCMWP. We propose a Lagrangean relaxation scheme. Lower and upper bound values are obtained by employing a sub gradient optimization algorithm. Furthermore an alternate location allocation heuristic is also suggested. The proposed solution procedures are tested on random test instances.


SET COVERING, SET PARTITIONING AND SET PACKING PROBLEM WITH LINEAR FRACTIONAL OBJECTIVE FUNCTION

Ratnesh Rajan Saxena and Rashmi Gupta

University of Delhi, Delhi, India

Abstract. Set Covering, Set Partitioning and Set Packing problems belong to the class of 0-1 integer programming problems that are NP-complete. In this paper an enumerative technique using Combinotorics is developed to solve these problems with non-linear objective function, in particular the fractional objective function. Most of the existing procedures such as cutting plane technique or branch and bound technique for solving these problems have their basis in simplex algorithm. The well known Breadth First Search (BFS) and Depth First Search (DFS) techniques of graph theory form the basis of the proposed algorithms. The enumeration procedure developed in the proposed algorithms is a modification of these techniques. The algorithms developed in this paper are supported by numerical examples.


TRAJECTORY REGRESSION MODEL FOR INDOOR PEDESTRIAN FLOW ANALYSIS ON BILLBOARD EVALUATION

Thomas Liebig

Fraunhofer IAIS, Augustin, Germany

University of Bonn, Bonn, Germany

Abstract. Over the last few years new measurement technology has revolutionized the performance measurement in outdoor advertising. A handful of pioneer countries trace personal mobility now via GPS devices, which allows for precise performance results of arbitrarily positioned outdoor poster campaigns. However, GPS technology has the drawback that it cannot be applied indoors due to signal loss. In Switzerland and Germany many valuable posters are situated in public buildings such as train stations or shopping malls and their evaluation is of high interest. In this paper we therefore present a new approach for the evaluation of mixed indoor-outdoor campaigns. Our approach consists of a pedestrian movement model denoting quantities and trajectories of the people in restricted spaces. The model is supported by empirical traffic observations and can be integrated into standard trajectory evaluation. Our approach has been implemented for 27 major train stations in Switzerland.


A COMPARISON MODEL OF DATA ENVELOPMENT ANALYSIS AND ANALYTICAL HIERARCHY PROCESS FOR THE VENDOR PERFORMANCE EVALUATION IN A RETAIL COMPANY

Neslihan Nese Aldikaçti

Flo Magazacilik ve Hizmetleri A.S., Istanbul, Turkey

Abstract. In this study we address the measurement of supplier efficiencies in a retail company. For that purpose, we suggest a two stage approach which employs both the Data Envelopment Analysis (DEA) and the Analytical Hierarchical Process (AHP) methods. In the first stage, we measure the efficiency of the suppliers with DEA method. Next, in the second stage, we use the AHP method to verify the results obtained with the first stage. As an illustrative example, we have applied the proposed approach in the supplier selection process of a retail company. We have observed that the results obtained by both DEA and AHP methods are consistent.


GENETIC ALGORITHM APPROACH FOR THE INVENTORY ROUTING PROBLEM WITH BACKLOGGING

Stella Sofianopoulou

University of Piraeus, Piraeus, Greece

Abstract. We consider a multi-period inventory-routing problem where a vendor serves multiple geographically dispersed customers who receive units of a single product from a depot with adequate supply. The class of problems arising from the combination of distribution and inventory management decisions is perhaps the most striking example of this concept and is known as the inventory routing problem (IRP). In this category of problems, the inventory routing problem with backlogs (IRPwB) deals with determining inventory level, backlogging and vehicle routing decisions from a single depot to a set of n customers over a specific number of time periods, using a fleet of homogenous vehicles. The aim is to minimize the average daily cost for the planning period, while ensuring that inventory level capacity constraints are not violated. We first develop an Integer Programming model to provide an accurate description of the problem and in a second phase a Genetic Algorithm (GA) with suitably designed genetic operators, is employed in order to obtain near optimal solutions.


SIMULATED ANNEALING ALGORITHM APPLIED FOR THE NESTING OF PARTS FABRICATED BY LAYER MANUFACTURING

Vassilis Dedoussis, Vassilios Canellidis and John Giannatsis

University of Piraeus, Piraeus, Greece

Abstract. The optimization of the build volume is one of the most important tasks encountered in the process planning phase of Layer Manufacturing (LM). The optimization is achieved via the dense nesting of parts, to be fabricated, on the LM machine platform, under certain geometrical constrains imposed by the LM technology used. The present work examines the utilization of a Simulated Annealing in conjunction with an effective placement rule, based on the notion of No-Fit Polygons, as a mean of optimizing the build volume of LM technologies that due to technical or quality reasons exclude the fabrication of a part on top of another, e.g. Stereo lithography. The reliability of the proposed methodology is demonstrated via a case study concerning representative “real-world” parts/objects with quite general free form geometry.


OPTIMAL SUPPLY CONTRACT DESIGN WITH MULTIPLE SUPPLIERS UNDER SUPPLY AND DEMAND RISKS

Atilla Yalçin and Christian Wolf

University of Paderborn, DS & OR Lab, Paderborn, Germany

Abstract. We consider a decision problem where a buyer has to decide how to design supply contracts with multiple suppliers. Each supply contract with a supplier consists of a fixed and flexible volume agreement which incurs different costs. The fixed part of the supply contract is linear pricing scheme with discounts whereas the flexible part is a supply option agreement. The buyer can choose to implement a single sourcing strategy or to run a multiple sourcing strategy with different supplier flexibilities. This is especially important in cases where supply and demand risks exist. We provide a stochastic linear programming model formulation and show how it can be implemented by Flop C++. Furthermore we give some numerical illustrations.


A ROBUST APPROACH FOR A MINIMUM POWER BROADCAST PROBLEM IN WIRELESS SENSOR NETWORKS

Nihat Engin Toklu and Roberto Montemanni

Istituto Dalle Molle di Studi sull’Intelligenza Artificiale, Manno, Switzerland

Abstract. The minimum power broadcasting problem is a well-known optimization problem dealing with the topology configuration of a wireless sensor network, where the positions of the network terminals are known, and the transmission power of each terminal has to be chosen, so that a coverage area is identified for each of them. The goal is to identify a connected topology, allowing a source terminal to transmit to all the other terminals, with the lowest possible total usage of transmission power. Given that the terminals are usually equipped with small batteries, the minimization of the total transmission power allows the network to survive longer. In this paper, we consider a minimum power broadcasting problem where the transmission powers required to reach one terminal from another are subject to uncertainty. The uncertainty is caused by realistic issues like unfriendly weather conditions. We propose a robust three-stage approach for this problem and present our experimental results.


OPTIMAL TIMING OF A MODULAR EXPANSION OF A WASTEWATER TREATMENT PLANT

Yuri Lawryshyn

University of Toronto, Toronto, ON Canada

Abstract. We consider a municipality faced with the question of how big to make their new wastewater treatment facility to meet the demand of 10% expected growth in the number of new connections. Previously, we developed a real options framework for determining optimal plant size and showed that the model takes on the form of an Asian option. Furthermore, it was shown that if the connection rate growths are closely correlated with the market growth, then the penalty costs associated with having insufficient capacity to treat the wastewater can be effectively hedged, significantly reducing overall expected costs. In this study, we utilize an approximate analytical solution and optimize the plant size of a staged / modular expansion with optimal timing.


EFFICIENT MULTIPLE OBJECTIVE NEURAL NETWORK MAPPING OF STATE-WIDE HIGH SCHOOL ACHIEVEMENT

Nina Kajiji

The NKD Group, Inc, USA

University of Rhode Island, USA

Gordon H. Dash, Jr.

College of Business Administration, University of Rhode Island, USA

Abstract. This paper focuses on the use of artificial intelligence to improve the accuracy of estimated metrics that are used in public policy planning to assess public school achievement in mathematics and the English arts. In this study we implement a truncated 2nd order translog production function in the form of a double-log model specification to explain variation in two Rhode Island state high school achievement indexes. The mathematics achievement index (MAI) and the state-wide three year English language arts index (ELAI). The results obtained by OLS estimation are compared to those generated by the application of a non-parametric multiple objective radial basis function (RBF) artificial neural network (ANN). The solution differences across the alternative model solutions produce a sharp focus on the need to disaggregate policy initiatives on the modeled prediction factors. For example, both models corroborate a long-held conventional wisdom by reporting elasticity estimates that infer a negative effect on mathematics achievement with percentage increase in the non-white population in the school system. However, the nonlinear mapping objective of the ANN amplifies this result to show that it is the increase in the interaction between the percentage of the non-white student population and those students who are eligible for a free or reduced lunch that is most responsible for the observed reduction in MAI scoring performance. Additional new insights are produced across other traditional explanatory factors.


USING SIMULATED ANNEALING FOR TERRITORIAL OPTIMIZATION

María B. Bernábe Loranca, David E. Pinto Avendaño and Rogelio G. Velázquez

Facultad de Ciencias de la Computación, Benemérita Universidad Autónoma de Puebla, Puebla, México

Abstract. A particular method for solving the problem of territorial optimization requires a classification process based on clustering in which multiple comparisons must be performed in order to fulfil an objective function that minimizes the distances among objects with the purpose of achieving geographical compactness. The computational complexity of this problem is known to be in the NP-complete category, therefore, we have tackle the problem by means of heuristic methods. We have selected the simulated annealing technique to be applied inside the clustering algorithm, because of its capability of finding high quality suboptimal solutions. In general, in this paper we present a mathematical model and a geographical clustering algorithm for solving aggregation, a particular feature that exists in every problem of the territorial kind. The algorithm is combinatorial and, therefore, it obtains approximate solutions throughout the execution of the simulated annealing algorithm. The experiment carried out needed a factorial statistical analysis in order to model the parameters of the employed heuristic. The dataset are of the census type, with a well defined spatial component and a vector of descriptive variables.


A TAYLOR SERIES APPROACH TO THE NUMERICAL ANALYSIS OF THE M/G/1//N QUEUE UNDER MULTIPLE VACATION POLICY OF THE SERVER

Fazia Rahmoune and Djamil Aissani

LAMOS Laboratory of Modelling & Optimization of Systems, University of Bejaia, Algeria

Abstract. This paper presents a functional approximation of the M/G/1//N queue under multiple vacation policy of the Server built on a Taylor series approximation. The work presented is a part of research project on numerical algorithms based on series expansions of Markov chains with finite state-space S. Numerical examples are carried out to illustrate the performance of the approach, while numerical bounds are provided for quantities from stochastic reliability models to optimize the preventive maintenance policy, after modelling by vacation queuing systems.


NETWORK MANAGEMENT CONSIDERATIONS FOR OPTICAL ACCESS NETWORK DESIGN

Cédric Hervet and Chardy Matthieu

Orange Labs, France

Abstract. Due to the emergence of bandwidth-requiring services, telecommunication operators are being compelled to renew their fix access network, most of them favouring the Fiber To The Home (FTTH) technology. For long, network design strategies have been driven by mere deployment CAPital EXpenditures (CAPEX). Today however, experience gathered from former networks management strongly pushes in favour of the consideration of many other sources of costs for the design of networks and operational deployment schemes. This paper focuses on the decision problem of the optimization FTTH networks under Operations, Administration and Maintenance (OA&M) considerations. Mixed integer formulations, extending the classical generalized integer flow model, are proposed for the modelling of this decision problem. Extensive numerical tests on real-life instances prove the efficiency of branch and bound solving approaches for such models. Assessment of the economical impact of OA&M considerations is also made. Routing problems show the practical relevance of this approach. Extensive tests have been performed on real-life instances in order to assess both the efficiency of branch and bound solving approaches on our models and the economic impact of the integration of such considerations on CAPEX costs.


THE DECISION SUPPORT SYSTEM FOR CRISIS MANAGEMENT

Sadko Mandžuka, Jasmina Pašagic Škrinjar and Vlado Cetl

University of Zagreb, Zagreb, Croatia

Abstract. The development of Knowledge-based Decision Support Systems for Crisis Management requires the management of complex information (Recourses, Geospatial information, Traffic etc). In some steps it is needed to solve certain tasks of Operations Research (Resource Allocation, Schedule manpower, Transport tasks, etc). This paper describes the design and implementation of a Knowledge-Based interactive spatial decision support system based on Operations Research and Geographic Information Systems (GIS) to offer a flexible interface for crisis management in City Zagreb. Preliminary analysis indicates that the use of such tools offers increasing efficiency of many crisis management problems.


AN IMPROVED GENETIC ALGORITHM FOR FEED MIX PROBLEM

Rosshairy Abd Rahman and Razamin Ramli

School of Quantitative Sciences, UUM College of Arts and Sciences, Universiti Utara Malaysia, Sintok Kedah, Malaysia

Abstract. Animal feed formulation or feed mix is a pillar in animal farming industry. It is important to satisfy animal needs and at the same time increase profit margin. Thus, this paper aims to discuss a Hybrid Genetic Algorithm (GA) as a new technique to obtain a feasible solution while considering existing issues in this area. In this study, GA is hybridized with Artificial Bee Colony (ABC) algorithm with annealing injection to increase the exploration and exploitation in basic GA algorithm. Repair operator is then included in order to cater for the strictness of ingredients restriction in this on-going study.


REAL OPTIONS VALUATION OF ABANDONED FARMLAND

Michi Nishihara

Graduate School of Economics, Osaka University, Osaka, Japan

Abstract. This paper models the decision-making process of an owner of abandoned farmland as follows. An opportunity that enables development for non-agricultural use comes as an exponential distribution. Before the arrival of the conversion opportunity, the owner has the option to cultivate the land with sunk costs at an arbitrary time. As soon as the land conversion opportunity arrives, the owner will immediately sell the land to a developer for residential or commercial use. This assumption is consistent with the fact that the price of land for non-agricultural use is much higher than that of land for agricultural use. To maintain brevity, I assume that cash flows from agricultural and non-agricultural land use follow bi-dimensional geometric Brownian motions.


PERFORMANCE-BASED MULTI-CRITERIA DECISION MAKING IN SERVICE SECTOR: ANALYTICAL HIERARCHY APPLICATION FOR PERFORMANCE EVALUATION

Arzu Alevcan Altan

Bahçesehir University, Istanbul, Turkey

Y. Esra Albayrak

Galatasaray University, Istanbul, Turkey

Abstract. The concept of performance evaluation, resources are limited in our world always has been important and will continue to be important. Performance measurement is becoming vital competitive situations. For this reason, adapt to changes, innovations and changing customer needs within the most responsive organizations to quickly implement world-level performance criteria may have. Services, regardless of a sale of goods or services in the past marketed to consumers or businesses need and desire to saturation, regardless of the actions could be defined. Therefore, the structure of service systems, created to meet customer needs. Services sector to grow and gain an important place in national economies, is one of the most important developments of recent years. Increasing competition in this area, much more effort for the survival of service companies has revealed the need to draw. These developments, depending on the performance of the service definition, measurement, and hence change the value of measuring success, have gained increasing importance. Encountering problems with performance measurement of service sector businesses, the classic problem-solving tools to eliminate these problems in addition to benefiting from the process analysis and management techniques. The aim of the study to assess and improve the performance of service systems, many factors are included to create a model.


RESOURCE PREFERENCE BASED IMPROVEMENT HEURISTICS FOR RESOURCE PORTFOLIO PROBLEM

Umut Besikci and Ümit Bilge

Bogaziçi University, Istanbul, Turkey

Gündüz Ulusoy

Sabanci University, Istanbul, Turkey

Abstract. The multi-project problem environment under consideration involves multiple-projects with activities having alternative execution modes, a general resource budget and a resource management policy that does not allow sharing of resources among projects. The multi-project scheduling model for this problem environment is called Resource Portfolio Problem. There are three basic conceptual problems in RPP: (i) determining the general resource capacities from the given general resource budget (general resource capacities determination); (ii) dedication of the general resource capacities to projects (resource dedication) and finally (iii) scheduling of individual projects with the given resource dedications. In this study, different preference based improvement heuristics are proposed for general resource capacities determination and resource dedication conceptual problems. For general resource capacities determination, the current general resource capacity values are changed according to the resource preferences such that the resulting capacity state would be more preferable. Similarly for resource dedication, resource dedication values of projects are changed according to the preferences of projects for resources such that the resulting resource dedication state would be more preferable. These two improvement heuristics separates and couples the conceptual problems. Different preference calculation methods are proposed employing Lagrangian relaxation and linear relaxation of MRCPSP formulation.


MODELLING VEHICLE ROUTING PROBLEMS WITH DELIVERIES AND PICKUPS MORE REALISTICALLY

Gábor Nagy, Said Salhi and Niaz Wassan

University of Kent, Kent, UK

Abstract. The vehicle routing problem with deliveries and pickups is the most important problem within reverse logistics. We review various models for this problem, including three which we believe to be more realistic but have been hitherto largely ignored by researchers. These are analysed in some detail. A reactive tabu search metaheuristic is proposed and the results analysed with a view to a better understanding of the problem structure.


A MULTIPLE CRITERIA DECISION MAKING APPROACH FOR LOCATION ANALYSIS: THE PRIMITIVE COGNITIVE NETWORK PROCESS

Kevin Kam Fung Yuen

Zirve University, Kizilhisar Campus, Gaziantep, Turkey

Abstract. Location selection is an essential decision problem for many industries such as hotel, tourism site, energy, school, warehouse, factory, stores, retails and estate. This paper proposes the Primitive Cognitive Network Process for the location selection problem considering multiple criteria. An illustrated example of hospital site selection demonstrates the usability of the proposed approach.


SCHEDULING JOBS ON IDENTICAL MACHINES WITH SPLIT AGREEMENT GRAPH

Mourad Boudhar

Faculty of Mathematics, USTHB University, Algiers, Algeria

Mohamed Bendraouche

Faculty of Sciences, Saad Dahleb University, Blida, Algeria

Abstract. We treat the following problem of scheduling with agreements: a set of jobs must be scheduled non-pre-emptively on two identical machines under the constraints that only some specific jobs can be scheduled simultaneously on different machines. These constraints are modelled by an agreement graph and the aim is to minimize the makespan. This problem is polynomial when the processing times are either 1 or 2. We prove that the latter becomes NP-hard when the release dates can take two distinct values 0 and r, even for split graphs. We also establish a polynomial result for split graphs when the jobs of the clique have unit processing times and those of the independent set are arbitrary.


EFFECT OF EDUCATIONAL TOOLS ON HPV VACCINE ACCEPTABILITY AMONG THAI ADOLESCENTS

Archin Songthap

Sirindhorn College of Public Health, Trang, Thailand

Abstract. This quasi-experimental study proposed to assess the changes in HPV vaccine acceptability, and knowledge and attitude regarding HPV, cervical cancer, and HPV vaccine after intervention among Thai adolescents. Two secondary schools under the Ministry of Education located in Bangkok which recruited male and female students aged between 12 – 15 years were selected, one school as an intervention group and the other as the control group. Of 194 and 164 students for the intervention and the control groups were obtained. Data collection was done between June and September, 2010 using self-administered questionnaire and educational tools, including fact sheet and multimedia concerning HPV, cervical cancer, and HPV vaccine. Statistics used for data analysis included frequency, percentage, mean, standard deviation, paired t-test, and McNemar test. The results revealed that after intervention, level of knowledge in the intervention group changed significantly from moderate to high level. On the other hand, level of knowledge in the control group was observed at moderate level before and after intervention. Most of the attitude in the intervention group increased significantly from neutral to positive after intervention whereas the attitude in the control group was neutral before and after intervention. After intervention, proportion of students in the intervention group who were willing to get vaccination rose from 24% to 47% whereas the willingness to get vaccination decreased from 26% and 24% in the control group. Greater changes in knowledge, attitude, and acceptability were observed in the intervention group than those in the control group after intervention. Therefore, health care provider should consider educational tools to introduce HPV vaccine for Thai adolescent appropriately.


FACTORS AFFECTING THE FIRST ATTEMPT ACHIEVEMENT OF THE STANDARD TEST FOR PUBLIC HEALTH STUDENTS IN THAILAND

Rungpesh Bunthose, P Bunthose, A Phalitnonkiat, B Tankimhong and P Khamnuengsitthi

Sirindhorn College of Public Health, Chonburi, Thailand

Abstract. This study aimed to verify relationship between students‟ grade point average (GPA), preparation, learning and teaching management, learning and teaching environment, and the first attempt achievement of the standard test, and to investigate factors affecting the first achievement of the standard test. Populations of the study were 234 students in the Higher Professional Certificate of Public Health Program. Research tool was a rating scale questionnaire with alpha coefficient at 0.87. Since the ethical consideration was not strongly enforced in the country and the research results were revealed as an overall not specified to any sample to protect the sample‟s rights. Therefore, this research design had only proposed and approved by research committee of the college. Tuition and suggestions had done for weeks before the standard test and data were collected immediately after the test. Analysis statistics were percentage, mean, standard deviation, Pearson's product moment correlation coefficient, and stepwise multiple regression. Research results found that most students had GPA at 2.29-3.00 (53.4 percent). They achieved the first attempt of the standard test at 93.20 percent. GPA and learning and teaching method were associated with the first attempt achievement of the standard test at 0.05 level. GPA and learning and teaching method were factors affecting the first attempt achievement of the standard test at 5.90 percent. Research results implied that the college should support learning and teaching management for students “experience. Learning and teaching should be developed consequently according to students” interests and aptitudes for higher GPA and learning outcomes advocating.


UTILIZING FUZZY AHP TO EVALUATE MAIN DIMENSIONS OF BUSINESS PROCESS REENGINEERING

AA Anvary Rostamy, F Bakhshi Takanlou, M Shaverdi and B Amiri

Tarbiat Modares University, Tehran, Iran

Islamic Azad University, Ilam Branch, Ilam, Iran

University of Tehran, Tehran, Iran

Abstract. The goal of business process reengineering (BPR) is to achieve dramatic improvements in business measures of performance by radically changing the process design. The aim of this study is developing a fuzzy model to evaluate of the BPR dimensions taking subjective judgments of decision makers into consideration. Proposed approach is based on Fuzzy Analytic Hierarchy Process (FAHP) method. Drawing on the four dimensions of BPR, this research first summarized the evaluation indexes synthesized from the literature relating to BPR. Then, for screening these subcategories, 11 subcategories fit for BPR evaluation were selected through expert questionnaires. FAHP method is used in determining the weights of the criteria by decision makers. Result of paper shows that organizational culture is the most important dimension of business process reengineering while BPR is implemented.


A P-CENTRE RELIABLE FACILITY LOCATION PROBLEM WITH A LIMITED SERVICE CAPACITY

Masoumeh Taslimi and Reza Tavakkoli-Moghaddam

College of Engineering, University of Tehran, Tehran, Iran

Abstract. In most cases, facility location decisions are highly expensive to change and managers usually prefer to prevent failures to occur or at least reduce the number of disruptions. A large number of researches are done on reliable facility location models during the last decade. This paper presents a developed model based on a reliable fixed-charge facility location problem, which is closer to real world problems, such as locating emergency systems. The proposed model is a conflicting bi-objective NP-hard problem that minimizes the maximum transportation cost and fixed cost simultaneously. In order to evaluate the verification of the model, a set of computational experiments are performed by LINGO 8.0.


A NEW SUPPLY CHAIN NETWORK DESIGN MODEL WITH RISK-POOLING, LEAD TIME AND MULTI-ECHELON INVENTORY UNDER DEMAND UNCERTAINTY

Reza Tavakkoli-Moghaddam and Fatemeh Forouzanfar

South Tehran Branch, Islamic Azad University, Tehran, Iran

Saeed Ebrahimnejad

Karaj Branch, Islamic Azad University, Karaj, Iran

Abstract. This paper addresses a problem of designing a multi-echelon supply chain with the single sourcing type and the related inventory systems, in which distribution centres are given. We also presents a novel mathematical model considering the risk-pooling, lead time, multi-echelon inventory under demand uncertainty, routing of vehicles from distribution centres (DCs) to customers in order to give services in a stochastic supply chain system, simultaneously. This problem is formulated as a non-linear integer programming model. Each customer demand is not known and it is under the control of the normal distribution. All the possible capacity levels are determined for distribution centres. The inventory existence of these centres can lead to a great success in reaching the risk-pooling advantage in order to overcome the variability of customer demands. The aim of this model is to determine the number of the located distribution centres, their locations and capacity levels, to allocate customers to distribution centres and distribution centres to suppliers. It is also to determine the net lead time of distribution centres and the inventory control decisions on the amount of products ordered and the amount of safety stocks at each opened DC. In addition, it is to determine the service time of every distribution centres and routing decisions, such as determination of vehicles‟ routes starting from an opened distribution centre to serve its allocated customers and returning to that distribution centre. All these are done in a way that the total system cost is minimized. Finally, the GAMS software is used to solve the presented model, and the computational results are illustrated.


OPTIMAL KEYWORD BIDDING IN SEARCH-BASED ADVERTISING WITH TARGET CLICK-THROUGH RATE

Baris Selcuk

Bahçesehir Üniversitesi, Istanbul, Turkey

Ozgur Ozluk

San Francisco State University, San Francisco, CA USA

Abstract. Search-based advertising has become very popular since it provides advertisers the ability to attract potential customers with measurable returns. In this type of advertising, advertisers bid on keywords to have an impact on their ad‟s placement, which in turn affects the response from potential customers. An advertiser must choose the right keywords and then bid correctly for each keyword in order to maximize the expected revenue or attain a certain level of exposure while keeping the daily costs in mind. In response to increasing need for analytical models that provide a guidance to advertisers, we construct and examine a deterministic optimization model that minimizes total expected advertising costs while satisfying a desired level of average click-through rate. We investigate the relationship between our problem and the well-known continuous nonlinear knapsack problem, and then solve the problem optimally by utilizing Karush-Kuhn-Tucker conditions. We present practical managerial insights based on the analysis of both a real-life data from a retailer and a hypothetical data.


JOINT RELIABILITY AND INVENTORY DECISIONS UNDER A SERVICE CONSTRAINT IN SPARE PARTS

Baris Selcuk and Semra Agrali

Bahçesehir Üniversitesi, Istanbul, Turkey

Abstract. Reliability and inventory levels of spare parts are major factors that determine the service level provided by original equipment manufacturing (OEM) companies. In general, decisions on reliability and stock levels are made separately in practice, and academic literature offers little guidance on how to jointly make these two decisions. In order to fill in the gap in the literature and provide guidance to OEM companies, we jointly model the reliability and inventory problems. We consider two service measures: fill rate and downtime. Our models minimize the holding and emergency shipment costs subject to a limited reliability improvement budget and a target service level. We develop an algorithm that considers reliability and inventory decisions simultaneously, test our solution approach on real-life and randomly generated data sets, and compare the results with an approach that considers reliability and inventory decisions sequentially. Numerical results show substantial benefits of integrating reliability and inventory decisions.


© ORLab Analytics Inc