Nihat Kasap

Minimizing makespan on a single machine subject to random breakdowns

We investigate optimal sequencing policies for the expected makespan problem with an unreliable machine, where jobs have to be reprocessed in their entirety if preemptions occur because of breakdowns. We identify a class of uptime distributions under which LPT minimizes expected makespan.

Provider selection and task allocation issues in networks with different QoS levels and all you can send pricing

We investigate an optimization problem a firm faces when acquiring network capacity from multiple providers. We define two types of tasks the firm performs using data networks, and show that the time, bandwidth and quality requirements of each type are quite different. We formulate the associated problem as a cost minimization problem subject to quality and capacity requirements and offer multiple solution approaches. We analyze how different prices, quality and task distribution affect the optimal behavior of the firm. We also implement Generalized Benders Decomposition to solve a relaxation of this problem in order to obtain a tight lower bound.

Augmented neural networks and problem structure-based heuristics for the bin-packing problem

In this article, we report on a research project where we applied augmented-neural-networks (AugNNs) approach for solving the classical bin-packing problem (BPP). AugNN is a metaheuristic that combines a priority rule heuristic with the iterative search approach of neural networks to generate good solutions fast. This is the first time this approach has been applied to the BPP. We also propose a decomposition approach for solving harder BPP, in which subproblems are solved using a combination of AugNN approach and heuristics that exploit the problem structure. We discuss the characteristics of problems on which such problem structure-based heuristics could be applied. We empirically show the effectiveness of the AugNN and the decomposition approach on many benchmark problems in the literature. For the 1210 benchmark problems tested, 917 problems were solved to optimality and the average gap between the obtained solution and the upper bound for all the problems was reduced to under 0.66% and computation time averaged below 33 s per problem. We also discuss the computational complexity of our approach.

Using heuristic algorithms for capacity leasing and task allocation issues in telecommunication networks under fuzzy quality of service constraints

Nowadays, every firm uses telecommunication networks in different amounts and ways in order to complete their daily operations. In this article, we investigate an optimisation problem that a firm faces when acquiring network capacity from a market in which there exist several network providers offering different pricing and quality of service (QoS) schemes. The QoS level guaranteed by network providers and the minimum quality level of service, which is needed for accomplishing the operations are denoted as fuzzy numbers in order to handle the non-deterministic nature of the telecommunication network environment. Interestingly, the mathematical formulation of the aforementioned problem leads to the special case of a well-known two-dimensional bin packing problem, which is famous for its computational complexity. We propose two different heuristic solution procedures that have the capability of solving the resulting nonlinear mixed integer programming model with fuzzy constraints. In conclusion, the efficiency of each algorithm is tested in several test instances to demonstrate the applicability of the methodology.

A bandwidth sourcing and task allocation model in telecommunications under stochastic QoS guarantees

Graphical abstractDisplay Omitted HighlightsBandwidth provider selection and task allocation problem is modeled from firms perfective at managerial level.Delay and jitter are considered as random variables and integrated into model via chance constraints.A novel heuristic algorithm and a tight lower bound method are proposed.Firms optimal strategies are investigated under different scenarios. In this study, we model bandwidth provider selection and task allocation problem as an expected cost minimization problem with stochastic constraints. Two important parameters of quality of service (QoS), namely delay and jitter are considered as random variables to capture stochastic nature of telecom network environment. As solution methodology, stochastic model is converted into its deterministic equivalent and then a novel heuristic algorithm is proposed to solve resulting non-linear mixed integer programming model. We analyze how different prices, quality and task distribution affect the optimal behavior of the firm. Finally, performance of solution procedure is tested by several randomly generated scenarios and by a relaxation of this problem as tight lower bound.

A fuzzy long-term investment planning model for a genco in a hybrid electricity market considering climate change impacts

We study the long-term generation capacity investment problem of an independent power generation company (GenCo) that functions in an environment where GenCos perform business with both bilateral contracts (BC) and transactions in the day-ahead market (DAM). A fuzzy mixed integer linear programming model with a fuzzy objective and fuzzy constraints is developed to incorporate the impacts of imprecision/uncertainty in the economic environment on the calculation of the optimal value of the GenCo’s objective function. In formulating the fuzzy objective function we also include the potential impacts of climate change on the energy output of hydroelectric power plants. In addition to formulating and solving the capacity planning/investment problem, we also performed scenario-based (sensitivity) analysis to explore how investment decisions of the GenCos change when fuzziness (tolerance) in the maximum energy output of hydroelectric units and/or drought expectation increases. The proposed model is novel and investigates the effects of factors like drought expectations of climate changes, hydroelectric power plant investments, and other power generation technology investment options.

Optimal Pricing Strategies for Capacity Leasing Based on Time and Volume Usage in Telecommunication Networks

In this study, we use a monopoly pricing model to examine the optimal pricing strategies for “pay-per-time”, “pay-per-volume” and “pay-per both time and volume” based leasing of data networks. Traditionally, network capacity distribution includes short/long term bandwidth and/or usage time leasing. Each consumer has a choice to select volume based, connection-time based or both volume and connection-time based pricing. When customers choose connection-time based pricing, their optimal behavior would be utilizing the bandwidth capacity fully, which can cause network to burst. Also, offering the pay-per-volume scheme to the consumer provides the advantage of leasing the excess capacity to other potential customers serving as network providers. However, volume-based strategies are decreasing the consumers’ interest and usage, because the optimal behaviors of the customers who choose the pay-per-volume pricing scheme generally encourages them to send only enough bytes for time-fixed tasks (for real time applications), causing quality of the task to decrease, which in turn creating an opportunity cost. Choosing pay-per time and volume hybridized pricing scheme allows customers to take advantages of both pricing strategies while decreasing (minimizing) the disadvantages of each, because consumers generally have both time-fixed and size-fixed task such as batch data transactions. However, such a complex pricing policy may confuse and frighten consumers. Therefore, in this study we examined the following two issues: (i) what (if any) are the benefits to the network provider of providing the time and volume hybridized pricing scheme? and (ii) would this offering schema make an impact on the market size? The main contribution of this study is to show that pay-per both time and volume pricing is a viable and often preferable alternative to the only time and/or only volume-based offerings for a large number of customers, and that judicious use of such pricing policy is profitable to the network provider.

Provider selection and task allocation in telecommunications with QoS degradation policy

The information age that we are living in is characterized by exponentially increasing needs and corresponding means to access, transmit and use data in a variety of business settings. Fast growing demand, which is translated to market opportunities, has led to the emergence of many new and well-established firms entering into the telecommunications market, resulting in a crowded, highly competitive business environment with numerous providers and carriers offering a wide range of data services. Today’s firms use telecommunication networks in a variety of ways to carry out their daily communications such as video conferencing, voice over IP and other data-intensive transmissions. In this paper, we report on a study in which we investigate a cost optimization problem that a firm encounters when acquiring network bandwidth from a telecommunication market that consists of many backbone providers offering different combinations of pricing policies and quality of service (QoS) levels. After the acquisition of network resources (bandwidth), firms allocate these resources to their daily data transmissions (tasks) according to the QoS requirement of the tasks. In an optimal allocation scheme, it is generally presumed that each task has to be assigned to a network resource which is capable of providing an equal or higher level of QoS than required by the task. However, it is shown with the proposed heuristic approach (presented herein) that QoS degradations during the allocation of tasks can lead to more favorable outcomes, especially when certain cost penalty policies are applied to the reduction of QoS requirements.

An analysis of the Gezi Park social movement tweets

Purpose Twitter usage during Gezi Park Protests, a significant large-scale connective action, is analyzed to reveal meaningful findings on individual and group tweeting characteristics. Subsequent to the Arab Spring in terms of its timing, the Gezi Park Protests began by the spread of news on construction plans to build a shopping mall at a public park in Taksim Square in Istanbul on May 26, 2013. Though started as a small-scale local protest, it emerged into a series of multi-regional social protests, also known as the Gezi Park demonstrations. The paper aims to discuss these issues. Design/methodology/approach The authors sought answers to three important research questions: whether Twitter usage is reflective of real life events, what Twitter is actually used for, and is Twitter usage contagious? The authors have collected streamed data from Twitter. As a research methodology, the authors followed social media analytics framework proposed by Fan and Gordon (2014), which included three consecutive processes; capturing, understanding, and presenting. An analysis of 54 million publicly available tweets and 3.5 million foursquare check-ins, which account to randomly selected 1 percent of all tweets and check-ins posted from Istanbul, Turkey between March and September 2013 are presented. Findings A perceived lack of sufficient media coverage on events taking place on the streets is believed to result in Turkish protestors’ use of Twitter as a medium to share and get information on ongoing and planned demonstrations, to learn the recent news, to participate in the debate, and to create local and global awareness. Research limitations/implications Data collection via streamed tweets comes with certain limitations. Twitter restricts data collection on publicly available tweets and only allows randomly selected 1 percent of all tweets posted from a specific region. Therefore, the authors’ data include only tweets of publicly available Twitter profiles. The generalizability of the findings should be regarded with concerning this limitation. Practical implications The authors conclude that Twitter was used mainly as a platform to exchange information to organize street demonstrations. Originality/value The authors conclude that Twitter usage reflected Street movements on a chronological level. Finally, the authors present that Twitter usage is contagious whereas tweeting is not necessarily.

A Hybrid Heuristic Approach to Provider Selection and Task Allocation Problem in Telecommunications with Varying QoS Levels

In this research we examine a cost minimization problem for a firm that uses telecommunication networks to carry out their daily operations by acquiring network capacity from multiple providers. We consider a network environment where the firm can acquire network capacity with different service qualities and price points. We formulate the associated optimization problem considering quality and capacity requirements and offer a solution approach based on Genetic Algorithm (GA). Our model considers the tradeoff between the capacity acquisition cost and opportunity cost that occurs when target transmission rates in real-time tasks fall below a desired level. We model the capacity and the loss probability requirements explicitly and formulate delay and jitter as level matching constraints. We analyze how different prices, quality and task distribution affect the optimal behavior of the firm. We also compare three GA based heuristics and comment on the suitability of the GA approach for resource selection and task allocation problems.