Sharif Digital Repository

In recent years the advent of internet and web has transformed the process of data collection and make organizational and environmental data more accessible. On the one hand, many organizations have realized that the knowledge in these huge databases is key to supporting the various organizational decisions. Particularly, the knowledge about customers from these databases is critical for the marketing function. But, much of this useful knowledge is hidden and untapped. On the other hand, the intense competition and increased choices available for customers have created new pressures on marketing decision-makers and there has emerged a need to manage customers in a long-term relationship.... 

Facility location decisions play a prominent role in strategic planning of many firms, companies and governmental organizations. Since in many real-world facility location problems, the data are subject to uncertainty, in this thesis, we consider the P-center problem under uncertainty of demand nodes. Using Bertsimas and Sim approach, we modeled the problem as a linear optimization model. Furthermore, we develop a tabu search algorithm to solve the problem. Finally we designed some experiments to adjust the parameters of tabu search algorithm. We presented the numerical result accordingly  

Success in forecasting and analyzing sales for a given good or service can mean the difference between profit and loss for an accounting period and, ultimately, the success or failure of the business itself. Therefore reliable prediction of sales becomes a very important task. This thesis presents a novel sales forecasting approach by integration of Genetic Fuzzy Systems (GFS) and Data Clustering to construct a sales forecasting expert system. At first, we use stepwise regression analysis (SRA) to determine factors which have most influence on sales. At the next stage we divide our raw data into k clusters by means of K-means algorithm. Finally, all clusters will be fed into independent... 

This research proposes a novel hybrid method for stock price index forecasting relying on technical analysis, the fundamental analysis of capital market, neural networks technology, and fuzzy logic. The new method shows a good performance comparing complicated and time-consuming forecasting methods. Several factors influence stock price; for example, an important one is the previous trend of stock price. Fortunately, by developing technical analysis and introducing the various indices of this method the maximum use of historical data is made in order to forecast future prices. Another kind of factors is macro-economic variables that their influence on the long-term trends of the stock... 

Nowadays, competition in supplying the requirement of the customer having become so important and curtail between the companies, so marketing plays an important role in the progress of the firms. Therefore, business and marketing concept is one the important tools to achieve their potential goals in the target markets. For Esfahan Steel Company because of their product’s quality, marketing is so significant. With respect to put the third forge will be put into operation perfectly in this company, the production will be increased and reach to 3.7 million tones per year. In this research, we develop a Marketing Plan for the products of this company in the international target market  

The objective of this study is to investigate the effects of brand familiarity and product type on consumer responses to cause-related marketing (CRM). The construct of involvement is used to categorize products. Besides, the general reaction of consumers to CRM campaigns in low-CRM-awareness situation is assessed. A classical experimental design with 202 respondents is used and results are analyzed using repeated-measures ANOVA and paired t-tests. The results show that how the interaction of brand familiarity and product type can affect the consumer responses to cause-related marketing campaigns. Brand familiarity and interaction effect of familiarity and product type proved influential on... 

The healthcare industry collects huge amounts of healthcare data which, unfortunately, are not“mined” to discover hidden information for effective decision making. Discovery of hidden patterns and relationships often goes nexploited. Advanced data mining techniques can help remedy this situation. The diagnosis of diseases is a vital and intricate job in medicine. The recognition of heart disease from diverse features or signs is a multi-layered problem that is not free from false assumptions and is frequently accompanied by impulsive effects. Thus the attempt to exploit knowledge and experience of several specialists and clinical screening data of patients composed in databases to assist the... 

In this research a joint location- inventory problem is analyzed. We assumed two kinds of inventories: fast-moving and slow-moving goods. These two kinds of goods can be observed in after sales service systems. In the real world fast moving goods are stocked in all dealers but for slow moving goods there is no force to stock in all dealers. Therefore an important decision in these systems is where and how much we should stock the slow moving goods and how each dealer can supplies its demand for slow moving goods. In this research we model this problem as a joint location-inventory model which locates some warehouses in the network to stock the slow moving goods and assigns the dealers to... 

One of the main applications of revenue management is in healthcare industry. By combining these two concepts, subjects such as patient allocation and scheduling and no-show rates were developed in the literature. Up until now, allocation and scheduling were considered separately in the literature. Also, no-show rates were predicted using regression models and data mining tools and these predictions were rarely used in scheduling systems. In this thesis, we combined allocation based on patients’ insurance type with scheduling based on patients’ no-show rates. Also, patients’ no-show rates were predicted and these rates were used in the proposed scheduling model. Firstly, the integrated... 

In this study, we consider a newsvendor problem. The newsvendor in addition to determine order quantities must also determine the selling price of each product. The demand of each product is of a stochastic nature and depends on the selling price. It is assumed that products are cross-elastic. We also assume that retailers want to apply mixed bundling strategy. It means that in addition to selling products separately, they sell them as bundles. Two types of two-steps algorithms (p-constant, q-constant) were proposed. First we applied these algorithms for newsvendor problem with 2 single products and 1 bundle. After demonstrating its efficiency, algorithms were developed for solving problems... 

In this research, the location-inventory problem is developed in order to design a distribution center considering customers behavior. We consider two groups of customers who have different behavior on purchase. The customers may order the product via internet and receive it at their location or go to the nearest store and buy the product at the store center. We model the problem as a multi-objective programming model. The objectives are maximizing the profit as well as maximizing the market share. The model locates the distribution centers (DC) and allocates the customers to DCs. Two heuristic algorithms developed to solve the model: non- dominated Sorting Genetic Algorithm (NSGA-II) and... 

Assortment planning considers the problem of product classification in order to maximize the profit by considering related constraints. Integrating the assortment planning with inventory management plays an essential role in retail optimization. The importance of integrating these problems is related to customer behavior in substituting the products in case of shortage. The customer checks the availability of products in order of his/her priorities and buys a product if available otherwise leaving the store without purchase if he/she cannot find a product according to his/her preferences. Customer preferences are modeled through the definition of consumer types, where a type is a ranking of... 

In this study hybrid classification models by combining genetic algorithm and classifiers such as neural network and decision tree are presented and efficiency of these models is tested on 5 different databases against other proposed models in this area. This comparison shows that the model obtained by combining genetic algorithm and neural network obtains better results than other models. Afterwards this model is used as a decision support system in diagnosis heart disease and in addition to determining the parameters of neural network such as the number of hidden layers and the number of neurons in each layer, efficient features in diagnosis heart disease are also determined. Among the... 

In this thesis, the problem of the hotel revenue management with the customer choice behaiver has been studied . In this model , each customer has a preference order among the set of products. He purchased or booked his products according to his ordered list of preferences. Because the high number of hotel products are assumed, each customer can only choices different fare classes and the list of the preference customers is made up of different fare classes. Two different approaches have been used to solve this problem . The first approach is The linear programming model in which α percent of the capacity is allocated to business customers and the remaining capacity to leisuer customers... 

In the aftermath of mass-casualty events, main resources may be overwhelmed by the sudden jump in demand. To ration these resources with considering the patients priority for help, mean triage. For prioritizing the patient for help, the limitation of resources should be considered. These resources can be different. This thesis at the first step investigates how the priority assignment can be done in a clearing system and what the potential benefits would be when there are same resources in the system. At the next step, the difference between resources is considered. So in this thesis, the problem of priority assignment of patients in an emergency system is modeled and analyzed at first with... 

In the research, the order acceptance and scheduling problem is integrated with pricing in order to maximize the net profit. Two mixed linear integer programming models have been proposed for the problem. A two level tabu search algorithm is proposed to solve the model. The suitable structure is determined for the algorithm based on numerical experiments. Based on experimental design results, the hybrid of random initial solutions and swap move in neighborhoods determination is the best structure for the algorithm. Numerical analysis shows the reasonable performance of proposed algorithm  

Pricing and Revenue Management in systems that offer services in subscription as well as pay-per-use format simultaneously is one of the most significant challenges in service companies. One of the main criteria in these systems is keeping the loyal customers and attracting more subscribers. Pricing and capacity allocation are decisions to be made in such systems considering customer’s behavior. In this research, we consider a service system with subscription and pay per use options to analyze the price and capacity allocation based on customer’s bahvior. The Loss Aversion theory is assumed for customer’s behavior and its effect on decisions is analyzed. A greedy algorithm as well as a smart... 

In this research we study multi-commodity inventory-location problem under disruption risk that is formulated to a nonlinear integer-programming model. The objective function of this model is minimization of inventory and location costs simultaneously under disruption risk. These disruptions may be caused by natural events such as earthquake, flood, fire and man-made sources like labor strike or terrorist attack. The structural design of a supply chain greatly affects how these kinds of disruption risks influence the network. In this regard, this study proposes a strategy to reduce the impacts of disruption risk by considering substitute commodity in the multi-commodity inventory-location... 

Consider a production system with a single product which is demanded by n different customer classes and each customer class has a Poisson demand with rate of λ_i for class i customer class. The important note is that different customer classes have different values for the system and they have different lost sales cost for the system. Therefore, the optimal decision is not necessarily satisfying the customer which arrived sooner. The main idea in this thesis is that for this problem which needs rationing in stock out situations, we can use pricing in line with rationing.
First, we have formulated the problem for the case of two customer classes and two available prices. We formulated... 

The main purpose of this thesis, is the scrutiny of social influences on products sales and prices. That is what probable effects can be imposed on the amount of sale and price of products as a result of customer's satisfaction or dissatisfaction. In this research pricing of a new product is considered which is produced and supplied through market in a monopolized manner for a specific time period. At the first period, people purchase the product given factors such as price, quality, etc. a part of whom, put comments about their experience regarding the product quality on social networks. In following periods, alongside with aforementioned factors, new purchasers also consider positive or...