Operations Research Proceedings 2015

Modern MIP solving software incorporates dozens of auxiliary algorithmic components for supporting the branch-and-bound search in finding and improving solutions and in strengthening the relaxation. Intuitively, a dynamic solving strategy with an appropriate emphasis on different solving components and strategies is desirable during the search process. We propose an adaptive solver behavior that dynamically reacts on transitions between the three typical phases of a MIP solving process: The first phase objective is to find a feasible solution. During the second phase, a sequence of incumbent solutions gets constructed until the incumbent is eventually optimal. Proving optimality is the central objective of the remaining third phase. Based on the MIP-solver SCIP, we demonstrate the usefulness of the phase concept both with an exact recognition of the optimality of a solution, and provide heuristic alternatives to make use of the concept in practice.

In this paper we present a novel extended formulation for the line planning problem that is based on what we call “configurations” of lines and frequencies. Configurations account for all possible options to provide a required transportation capacity on an infrastructure edge. The proposed configuration model is strong in the sense that it implies several facet-defining inequalities for the standard model: set cover, symmetric band, MIR, and multicover inequalities. These theoretical findings can be confirmed in computational results. Further, we show how this concept can be generalized to define configurations for subsets of edges; the generalized model implies additional inequalities from the line planning literature.

This article provides an overview of the author’s dissertation (Berthold, Heuristic algorithms in global MINLP solvers, 2014, [4]). We study heuristic algorithms that are tightly integrated within global MINLP solvers and analyze their impact on the overall solution process. This comprises generalizations of primal heuristics for MIP towards MINLP as well as novel ideas for MINLP primal heuristics and for heuristic algorithms to take branching decisions and to collect global information in MIP.

We consider the FIFO Stack-Up problem which arises in delivery industry, where bins have to be stacked-up from conveyor belts onto pallets. Given k sequences $$q_1, \ldots , q_k$$ of labeled bins and a positive integer p. The goal is to stack-up the bins by iteratively removing the first bin of one of the k sequences and put it onto a pallet located at one of p stack-up places. Each of these pallets has to contain bins of only one label, bins of different labels have to be placed on different pallets. After all bins of one label have been removed from the given sequences, the corresponding stack-up place becomes available for a pallet of bins of another label. The FIFO Stack-Up problem is computational intractable (Gurski et al., Math. Methods Oper. Res., [3], ACM Comput. Res. Repos. (CoRR), 2013, [4]). In this paper we consider two linear programming models for the problem and compare the running times of our models for randomly generated sequences using GLPK and CPLEX solvers. We also draw comparisons with a breadth first search solution for the problem (Gurski et al.,Modelling, Computation and Optimization in Information Systems and Management Sciences, 2015, [7]).

In this paper a class of network optimization problems is discussed. It is assumed that a partial solution can be formed in the first stage, when the arc costs are precisely known, and completed optimally in the second stage, after a true second-stage cost scenario occurs. The robust min-max criterion is used to compute an optimal solution. Several complexity results for the interval and discrete uncertainty representations are provided.

This paper deals with the emergency service system design, where not only the disutility of an average user is minimized, but also the disutility of the worst situated user must be considered. Optimization of the average user disutility is often related to the weighted p-median problem. To cope with both objectives, we have suggested a composed method. In the first phase, the disutility of the worst situated user is minimized. The second phase is based on the min-sum approach to optimize the average user’s disutility. The result of the first phase is used here to reduce the model size. We focus on effective usage of the reduction and explore the possibility of a trade-off between a little loss of optimality and computational time.

We consider the one-dimensional skiving stock problem (SSP) which is strongly related to the dual bin-packing problem in literature. In the classical formulation, different (small) item lengths and corresponding availabilities are given. We aim at maximizing the number of objects with a certain minimum length that can be constructed by connecting the items on hand. Such computations are of high interest in many real world application, e.g. in industrial recycling processes, wireless communications and politico-economic questions. For this optimization problem, we give a short introduction by outlining different modelling approaches, particularly the pattern-based standard model, and mentioning their relationships. Since the SSP is known to be NP-hard a common solution approach consists in solving an LP-based relaxation and the application of (appropriate) heuristics. Practical experience and computational simulations have shown that there is only a small difference (called gap) between the optimal objective values of the relaxation and the SSP itself. In this paper, we will present some new results and improved upper bounds for the gap of the SSP that are based on the theory of residual instances of the skiving stock problem.

We present a new approach to create instances with large performance ratio, i.e., worst-case instances, of common heuristics for the two-dimensional Strip Packing Problem. The idea of this new approach is to optimise the width and the height of all items aiming to find an instance with maximal performance ratio with respect to the considered heuristic. Therefore, we model the pattern obtained by the heuristic as a solution of an ILP problem and merge this model with a Padberg-type model of the two-dimensional Strip Packing Problem. In fact, the composed model allows to compute the absolute worst-case performance ratio of the heuristic with respect to a limited number of items. We apply the new approach for the Next-Fit Decreasing-Height, the First-Fit Decreasing-Height, and the Best-Fit Decreasing-Height heuristic. Furthermore, we provide an opportunity to use this idea to create worst-case instances for lower bounds.

In the Maximum Scatter Traveling Salesman Problem the objective is to find a tour that maximizes the shortest distance between any two consecutive nodes. This model can be applied to manufacturing processes, particularly laser melting processes. We extend an algorithm by Arkin et al. that yields optimal solutions for nodes on a line to a regular ($$m \times n$$)-grid. The new algorithm $$\textsc {Weave}(m,n)$$ takes linear time to compute an optimal tour in some cases. It is asymptotically optimal and a ($$\frac{\sqrt{10}}{5}$$)-approximation for the ($$3\times 4$$)-grid, which is the worst case.

We consider the 2D rectangular strip packing problem without rotation. A relaxation of that problem is the 1D horizontal bar relaxation, the LP relaxation of the 1D binary cutting stock problem. To represent a solution of the strip packing problem, a solution of the horizontal bar relaxation has to satisfy, among others, the vertical contiguous condition. To strengthen the bar relaxation with respect to that vertical contiguity, we investigate a cutting plane approach. Furthermore, a solution of the bar relaxation must ensure constant location of items. Because 1D cutting patterns do not provide any information about the location of the items contained, we investigate methods to provide 2D feasibility of patterns in the column generation and cutting plane process, respectively. Some computational results are also reported.

We consider the following capital budgeting problem. A firm is given a set of investment opportunities $$X=\{x_1,\ldots ,x_n\}$$X={x1,…,xn} and a number m of portfolios. Every investment $$x_i$$xi, $$1\le i\le n$$1≤i≤n, has a return of $$r_i$$ri and a price of $$p_{i}$$pi. Further for every portfolio j there is capacity $$c_j$$cj. The task is to choose m disjoint portfolios $$X'_1,\ldots , X'_m$$X1′,…,Xm′ from X such that for every $$1\le j\le m$$1≤j≤m the prices in $$X'_j$$Xj′ do not exceed the capacity $$c_j$$cj and the total return of this selection is maximized. From a computational point of view this problem is intractable, even for $$m=1$$m=1 [8]. Since the problem is defined on inputs of various informations, in this paper we consider the fixed-parameter tractability for several parameterized versions of the problem. For a lot of small parameter values we obtain efficient solutions for the partitioning capital budgeting problem. We also consider the connection to pseudo-polynomial algorithms.

We consider the capacitated competitive facility location problem (CCFLP) where two competing firms open facilities to maximize their profits obtained from customer service. The decision making process is organized as a Stackelberg game. Both the set of candidate sites where firms may open facilities and the set of customers are finite. The customer demands are known, and the total demand covered by each of the facilities can not exceed its capacity. We propose the upper bound for the leader’s objective function based on solving of the estimating MIP.

With the expected uptake of electric vehicles in the near future, we are likely to observe overloading in the local distribution networks more frequently. Such development suggests that a congestion management protocol will be a crucial component of future technological innovations in low voltage networks. An important property of a suitable network capacity management protocol is to balance network efficiency and fairness requirements. Assuming a stochastic model, we study the proportional fairness (PF) protocol managing the network capacity in charging of electric vehicles. We explore the onset of congestion by analysing the critical arrival rate, i.e. the largest possible vehicle arrival rate that can still be fully satisfied by the network. We compare the proportionally fair management protocol with the max-flow (MF) management protocol. By numerical simulations on realistic networks, we show that proportional fairness leads not only to more equitable distribution of power allocations, but it can also serve slightly larger arrival rate of vehicles. We consider simplified setup, where the power allocations are dependent on the occupation of network nodes, but they are independent of the exact number of vehicles, and to validate numerical results, we analyse the critical arrival rate on a network with two edges, where the optimal power allocations can be calculated analytically.

The prize-collecting Steiner tree (PCST) problem is a broadly studied problem in combinatorial optimization. It has been used to model several real world problems related to utility networks. More recently, researchers have started using PCSTs to study biological networks. Biological networks are typically very large in size. This can create a considerable challenge for the available PCST solving methods. Taking this fact into account, we have developed methods for the PCST that efficiently scale up to large biological network instances. Namely, we have devised a heuristic method based on the Minimum Spanning Tree and a matheuristic method composed of a heuristic clustering phase and a solution phase. In this work, we provide a performance comparison for these methods by testing them on large gene interaction networks. Experimental results are reported for the methods, including running times and objective values of the solutions.

We consider the problem of allocating database segments to the locations of a content distribution network (CDN). Many other decisions such as server location, query routing, user assignment and network topology are also addressed simultaneously. We consider the problem with the backbone (server network) being restricted to be a tree. Although it is an extremely hard problem to solve in its original form, prior information on the segment allocation, server location, and the tree backbone, reduces the original problem to a simple assignment problem, and therefore, indicates the suitability of a decomposition approach. We formulate the problem and develop a Benders decomposition approach. Due to the hardness of the master problem, we solve it heuristically to obtain a reasonable upper bound on the original problem in a short period of time. The success of the algorithm is particularly significant in large problems, for which CPLEX struggles to obtain even a feasible integer solution.

Magnetic rulers for measuring systems are either based on incremental or absolute measuring methods. Incremental methods need to initialize a measurement cycle at a reference point. From there, the position is determined by counting increments of a periodic graduation. Absolute methods do not need reference points, since the position can be read directly from the ruler. In the state of the art approach the absolute position on the ruler is encoded using two tracks with different graduation. To use only one track for position encoding in absolute measuring a pattern of trapezoidal magnetic areas is considered instead of the common rectangular ones. We present a mixed integer programming model for an optimal placement of the trapezoidal magnetic areas to obtain the longest possible ruler under constraints conditioned by production techniques, physical limits as well as mathematical approximation of the magnetic field.

Given a set of departments, two rows with a common left origin and pairwise connectivities between the departments, the Space-Free Double-Row Facility Layout Problem (SF-DRFLP) looks for permutations of the departments in both rows such that the weighted sum of the center-to-center distances between all pairs of departments is minimized. In this paper we present a new mixed-integer linear programming formulation for the (SF-DRFLP) with given assignment of the departments to both rows that combines distance and betweenness variables. Furthermore, we analyze the combinatorial structure of optimal solutions, which allows us to prove a certain balancing condition. We then use this formulation in an enumeration scheme for solving the (SF-DRFLP). Indeed, we test all possible row assignments, where some assignments are excluded by our new combinatorial investigations. This approach allows us to solve (SF-DRFLP) instances with up to 16 departments to optimality for the first time.

To improve competitiveness, small and mid-sized carriers ally in horizontal carrier coalitions for request exchange. A crucial aspect for the long-term viability and stability of coalitions is a fair cost allocation among the agents. Despite of the long computing time, the well-known Shapley value has been used as a scheme for cost allocation. The contribution of this paper lies on the development of a suitable sampling procedure that approximates the Shapley value applied to cost allocations for the collaborative traveling salesman problem with time windows. A computational study identifies the deviation of the values generated by the proposed sampling procedures from the actual Shapley value.

In collaborative transportation planning, independent forwarders align their transportation plans by exchanging requests within a horizontal coalition. The goal of the coalition members is to increase their profitability and flexibility in competitive markets with high demand fluctuations. In recent publications, it is assumed that each request can be fulfilled by any coalition member. However, in practice some requests are prohibited to be forwarded due to contractual agreements. These requests are known as compulsory requests. The contribution of this paper is to identify the increase of costs caused by compulsory requests of a collaborative pickup and delivery transportation planning problem. To analyze the impact of compulsory requests, an existing column generation-based heuristic with two solution strategies for handling compulsory requests is applied and investigated.

In the well-known Vehicle Routing Problem (VRP), customer demands from one or more depots are to be distributed via a fleet of vehicles. Various objectives of the problem are considered in literature, including minimization of the total distance/time traversed by the fleet during distribution, the total cost of vehicle usage, or minimizing the maximum tour length/time. In this study, we consider a multi-compartment VRP with incompatible products for the daily solution of a livestock feed distribution network, where each livestock farm requests one type of feed from a single depot, and the vehicles have several compartments. The objective is to minimize the total cost of distribution. Although VRP is a well-studied problem in literature, multi-compartment VRP is considered only by few authors, and our problem differs from the existing ones due to special operational constraints imposed by the situation on hand. We formulate a basic mathematical model for the problem and present possible extensions. We design a computational experiment for testing the effects of uncontrollable parameters over model performance on a commercial solver and report the results. The proposed model can easily be adapted to other distribution networks such as food and fuel/chemicals.

We consider the problem of finding the optimal routing of a single vehicle that starts its route from a depot and delivers a product to N customers that are served according to a particular sequence. The vehicle during its route can return to the depot for restocking. The demands of the customers are random variables with known distributions. The actual demand of each customer is revealed as soon as the vehicle visits the customer’s site. It is permissible to satisfy fully or to satisfy partially or not to satisfy the demand of a customer. The cost structure includes travel costs between consecutive customers, travel costs between the customers and the depot and penalty costs if a customer’s demand is not satisfied or if it is satisfied partially. A dynamic programming algorithm is developed for the determination of the optimal routing policy. It is shown that the optimal routing policy has a specific threshold-type structure.

The container relocation problem (CRP) is concerned with clearing out a single yard-bay which contains a fixed number of containers each following a given pickup order so as to minimize the total number of relocations made during their retrieval process. In this work, we consider an extension of the CRP where containers are both received and retrieved at a single yard-bay named Dynamic Container Relocation Problem (DCRP). The arrival and departure sequences of containers are assumed to be known in advance. A tabu search based heuristic approach is proposed to solve the DCRP. Computational experiments are performed on an extensive set of randomly generated test instances from the literature. Our results show that the proposed algorithm is efficient and yields promising outcomes.

We consider the allocation of a running time supplement to a railway timetable. Previously, Vekas et al. developed a stochastic programming model. In this paper, their optimization model is improved by adding bound constraints on the supplements. It is shown that the probability of delays decreases when using the proposed model. In addition, an effective L-shaped algorithm is presented.

To cope with growing passenger demand and to provide better services, increasing the number of trains is desired for certain lines. Maximum possible throughput of a line is often limited due to the limited number of platforms at the terminal, which was the case with two bullet-train lines originating from Tokyo. In these lines, there exists an intermediate station in the close vicinity of the terminal. This paper proposes a network-based optimization model to analyze the throughput of a line when turn-backs at the adjacent station are introduced to increase the line capacity.

The introduction of the Sulphur Emission Control Areas is expected to lead to increasing costs in maritime transportation with a high impact on Short Sea Shipping in the North Sea and the Baltic Sea. As a consequence a modal shift in multimodal container transport might occur towards road and rail freight transport, which is related to negative ecological effects. Simultaneously, a downside risk of the North Range harbours exist in international container transport due to their competition with the south European harbours. Based on a case study it will be analysed how and to what extent the new frame conditions influence the container business and what consequences can be expected.

The quantification of the differences of two sets of vehicle routes is addressed. We present procedures to compute indicators quantifying differences in the tour composition among both sets. Furthermore, we propose two different ideas to compare sequencing decisions related to the routing sub-problem in vehicle routing. Finally, we combine clustering and sequencing decision comparison procedures in order to quantify differences of the vehicle route sets.

For container transport from an overseas port to the final customer destination, the decision whether the transport should be by direct truck or by an intermodal option, including train and truck, is rather relevant. An optimization problem minimizing transportation costs and delay costs is developed. For unknown container arrival times, deterministic transportation times and a predefined container delivery date, the optimal transportation mode is discussed. For a predefined train departure time, the relationship between unplanned truck delivery if arrival time is after train departure and expected delay for later train departure times is shown. For the intermodal option, an optimization problem minimizing transportation and expected delay costs is solved. Results show that there is a certain delivery date threshold above which the intermodal option becomes optimal. A counter-intuitive finding from the numerical results indicates that better information quality does not necessarily favour the intermodal option. Especially for low delivery date values, the additional cost to be paid for the intermodal option increases with better information quality.

Motorail transportation covers the loading of various types of vehicles onto transportation wagons. The detailed treatment of realistic technical and legal constraints by integer linear programming models is challenging especially in terms of computation times. This paper considers decision support for the loading process at motorail terminals with the goal of speeding up the entire loading process while guaranteeing the feasibility of the proposed loading plan at all times. Specially tailored integer programming formulations are proposed and their suitability for real-world use is evaluated by means of a case study.

Home health care providers face a complex routing and scheduling task to plan their services because their clients stay at their own homes. As the solution of this task may be inefficient or infeasible for a subsequent planning period, a new optimization is inevitable at the end of each period. The consideration of continuity in multi-period planning by avoiding extensive changes between periods is essential to ensure client and nurse satisfaction. To address this issue, we consider the home health care problem in a rolling planning horizon. Our heuristic solution method determines a new plan while preserving the continuity between periods. Since there are many possibilities to quantify continuity, we compare different measures and show their impact on the solutions.

In this paper we propose key figures to compare two different partitions of street maps. The first partition is used in navigation and reduce the cutting edges in the partition. The second is dependent on a new standard for the transmission of navigational data and has a rectangular shape. We build and analyze the relationship of the partitions with real map data and present first results.

OD matrix is an important input parameter for a large number of optimization problems especially in the public transportation. Traditional approaches of obtaining OD matrix, such as surveys, could not enable us to obtain comprehensive and complex data on passengers and their journeys. In cases where passengers in transportation use smart cards, we can obtain more accurate data about the passengers journeys even in cases where these data are incomplete. In this contribution we present a trip-chaining method to obtain passenger journeys from smart card transactions data. Using these transactions data with the combination of data from other sources such as street maps, timetable and bus line routes, we are able to obtain origins and destinations of passenger journeys and also information about the changes between the lines on the passengers journey. Designed approach is verified on the case of the Zilina municipality with a data set with real passengers smart card transactions for a period of one week. Obtained OD matrix is later used as the input for the solving of tariff zones partitioning problem in the Zilina municipality area and these results are also presented in this paper.

Fuel prices on the rise and ambitious goals in environment protection make it increasingly necessary to change for modern and sustainable powertrain technologies. This trend also affects the transportation sector, and electric buses with stationary charging technology grow in popularity. Their launch however is still costly, and an optimal choice of the charging stations locations is crucial. In our paper we present a mixed integer model that determines an optimal solution concerning the investment costs for a single bus line. It is mainly constrained by an energy balance. Hence, energy consumption on the driven paths and of auxiliary consumers have to be considered as well as holding times at bus stops and thus the potentially recharged amount of electric energy. Additionally, we take account of service life preservation of the batteries as well as beneficial existing infrastructure and constructional restrictions. We give an overview of our results obtained from real world data of the bus network of Mannheim. In our tests we consider different scenarios regarding passenger volume, traffic density and further factors.

This paper presents a model for railway crew scheduling problems dealing with attendance rates for conductors. Afterwards we discuss a hybrid solution approach for these kind of problems. This approach consists of a column generation framework using genetic algorithm to solve the pricing problem. Based on a real-world instance, we compare our hybrid solution approach with the enumeration approach with respect to resulting total costs and computation time.

In this paper we present a method, which is able to compare the use of pure combustion vehicles with hybrid electric vehicles and plug-in hybrid electric vehicles for end customer deliveries. Benchmark instances representing typical delivery areas for small package shipping companies are introduced. For small instance sizes, we are able to generate exact solutions with standard mixed-integer program solver software. In contrast to the exact approach, our heuristic allows us to solve practical instance sizes.

In this work we address the concept of robust efficiency of resource schedules in public bus transport and airline industry, dealing with the competing objectives of cost-efficiency and robustness. Generalizing the findings from two research projects we provide techniques that lead to an improvement of the pareto-front between robustness and cost-efficiency. These techniques include the improvement of scheduling and optimization approaches as well as a refinement of delay prediction models enabling a robustness evaluation closer to reality. Additionally, problem characteristics in public transport and airline network topologies and their influence on the degree of freedom for robust resource scheduling and dispatching strategies are examined.

In this work we introduce the vehicle routing problem with multiple pauses, where the fleet is heterogeneous in terms of capacity and drivers availability. Each shift has a time interval when the driver is available and a set of breaks that needs to be scheduled in the route during this shift. The objective is to minimize the number of vehicles and the travel distance. To tackle large instances, we develop a three-phase local search algorithm taking multiple breaks into account by introducing an ejection pool and randomized variable neighborhood descent as local improvement procedure. For effective break scheduling, we develop a special dynamic programming routine. Computational experiments are done on the data set provided by a delivery company situated in Novosibirsk, Russia. The instances contain 1000 customers and 30 vehicles. Experiments show effectiveness of our algorithm. It substantially reduces the fleet and travel distance.

Vertical flight planning of commercial aircrafts can be formulated as a Mixed-Integer Linear Programming (MILP) problem and solved with branch-and-cut based solvers. For fuel-optimal profiles, speed and altitude must be assigned to the corresponding segments in such a way that the fuel consumed throughout the flight is minimized. Information about the fuel consumption of an aircraft is normally given by the aircraft manufacturers as a black box function, where data is only available on a grid points depending on speed, altitude and weight. Hence, some interpolation technique must be used to adequate this data to the model. Using piecewise linear interpolants for this purpose is suitable for the MILP approach but computationally expensive, since it introduces a significant amount of binary variables. The aim of this work is to investigate reductions of the computation times by using locally optimal solutions as initial solutions for a MILP model which is, thereafter solved to global optimality. Numerical results on test instances are presented.

To guarantee a safe journey for each aircraft, air traffic controllers make sure that separation minima are maintained. To prevent controller overburdening, each controller team is responsible for one confined sector. Furthermore, sectors limit the number of flights entering each hour. Compliance to all sector and airport capacity constraints in daily business is ensured by EUROCONTROL’s Network Management. This function balances the flights’ demand of airspace with available capacity by re-allocating departure timeslots. However, when minimum separation between two aircraft may become compromised, a conflict occurs. Conflicts are solved by controllers who provide pilots with instructions to maintain separation. Hence, conflicts increase controller workload and thus tighten sector capacity. The aim of this work is a prevention of actual conflicts by strategic deconfliction. Strategic conflicts refer to planned trajectories which violate separation minima in any future point in space and time. Mimicking a future Network Management, departure times are re-allocated to reduce the number of strategic conflicts while satisfying both sector and airport capacity constraints. As a basis for the deconfliction, actual datasets of planned trajectories, sector bounds and airport features are aggregated to model the European Air Traffic Flow Management. The allocation problem of departure timeslots is formulated as a Quadratic Binary Problem with linear delay costs, quadratic conflict costs and linear constraints. In an optimal solution, all strategic conflicts are solved. Finally, a trade-off between conflict reduction and delay is performed.

Timetabling and capacity planning of railway transport faces ever-growing challenges. Due to the high number of different influences on capacity, timetable optimization in the railway network cannot be efficiently handled by manual effort. The software system TAKT is a state-of-the-art realization, which allows to compute automatically strictly synchronized and conflict-free timetables for very large railway networks. The complexity increases significantly in the consideration of single tracks and highly frequented main railway stations which may also have a extensive track layout. This work shows how the complexity of the timetables process can be reduced by ignoring selected minimum headway constraints. As a result, timetables with possible conflicts in those covered regions will be computed. Consequently, there is the need for efficient algorithms and its corresponding conjunction to solve the remaining conflicts by detecting alternative stopping positions and routes within a main railway station and the optimized selection.

Multi-stage, multi-criteria discrete decision processes under risk are considered and Bellmans principle of optimality is applied. Quasi-hierarchy of multi-period criteria is determined by the decision maker. The aim of the paper is to propose an algorithm to solve quasi-hierarchical problem according to decision maker’s requirements. The process of obtaining the final solution is interactive.

In times of planned obsolescence the demand for sustainability keeps growing. Ideally, a technical system is highly reliable, without failures and down times due to fast wear of single components. At the same time, maintenance should preferably be limited to pre-defined time intervals. Dispersion of load between multiple components can increase a system’s reliability and thus its availability inbetween maintenance points. However, this also results in higher investment costs and additional efforts due to higher complexity. Given a specific load profile and resulting wear of components, it is often unclear which system structure is the optimal one. Technical Operations Research (TOR) finds an optimal structure balancing availability and effort. We present our approach by designing a hydrostatic transmission system.

VHF radars are suitable in some air surveillance applications, due to their cost-effectiveness and the fact that radar cross section of an aircraft is larger at VHF band than at higher frequencies, making detection easier. To ensure coverage of all ranges and velocities of interest, contemporary VHF radars utilize a complex waveform. We formulate the design of this waveform as a multiobjective optimization problem, with signal-to-noise ratio (SNR), Doppler visibility and Doppler resolution as objectives, which should be maximized. We show that the objectives are in conflict and use a particular example to explore the Pareto frontier (PF) for the problem. We find that reasonable tradeoff can be made between SNR and Doppler visibility, leading to an idea of multiple modes of operation, selectable at run time. We conclude that this subject is worth of further investigation, and that finding an efficient method for determining the PF would facilitate further research.

Dealing with ill-defined optimization problems, where the actual values of the input parameters are unknown or not directly measurable, is generally not an easy task. In order to enhance the robustness of the final solutions, we propose in the current paper a hybrid metaheuristic approach that incorporates a sampling-based simulation module. Empirical application to the classical mean-variance portfolio optimization problem, which is known to be extremely sensitive to noises in asset means, is provided through a genetic algorithm solver. Results of the proposed approach are compared with that specified by the baseline worst-case scenario and the two approaches of stochastic programming and robust optimization.

We consider the base stations location problem with sharing. New operator arrives the market and competes with an existing network of another operator. Latter one can share BS cites with new operator, receiving a rent payment from him. We propose new model of realistic clients behavior and formulate the problem as a nonlinear integer programming problem. We propose a fast tabu search heuristic for this problem and provide some computational results.

Finding an appropriate IT security strategy by implementing the right security safeguards is a challenging task. Many organizations try to address this problem by obtaining an IT security certificate from a recognized standards organization. However, in many cases the requirements of a standard are too extensive to be implemented, particularly by smaller organizations. But the knowledge contained in a security standard may still be used to improve security. Organizations that have an interest in security but not in a certificate, face the challenge of utilizing this knowledge and selecting appropriate safeguards from the given standard. To solve this problem, a new robust optimization model to determine an optimal safeguard configuration is proposed. By incorporating multiple threat scenarios, obtained solutions are robust against uncertain security threats.

Critical Infrastructure Protection (CIP) is a challenging operation for all involved organisations: Authorities, critical infrastructure providers and even policy makers. Integrated risk management is required to keep risks as low as possible and well-developed crisis management helps to mitigate the effects of events that have occurred. Achieving the right balance is difficult especially for anthropogenic threats such as terrorist threats, which are difficult to assess with normative risk management approaches. In May 2015, the TH Köln executed two exercises to address risk and crisis management in case of terrorist threats. The exercises were embedded in the research project RiKoV, which was being funded by the German Federal Ministry of Research and Education.

Critical infrastructure protection represents one of the main challenges for decision makers today. This paper focuses on rail-based public transport and on the interaction of the station layout with passenger flows. Recurring patterns and accumulation points with high passenger densities are of great importance for an analysis since they represent e.g. critical areas for surveillance and tracking and further security implementations. An agent-based model is developed for crowd behavior in railway stations. For the analysis, we apply the methodology of data farming, an iterative, data-driven analysis process similar to the design of simulation experiments. It uses experimental designs to scan the parameter space of the model and analyses the data of the simulation runs with methods stemming from statistics and data mining. With its help, critical parameter constellations can be identified and investigated in detail.

In this paper, the frequency of trains in the Munich subway network is analyzed. Using influence diagrams the stations and edges in the network that are most vulnerable to catastrophic attacks are determined. Upon obtaining the number of trains in each station at a certain moment in time, the most vulnerable stations will be automatically identified. This process is discrete in time, and various existing train schedules available to the general public are considered. Considering each schedule, the gain and the cost of destroying a station is calculated. Based on utility values for each station representing the difference between the gain and the cost, an influence diagram decides which stations are most vulnerable to attacks.

This study proposes two approximation methods to define the coverage probability in ambulance location problems based on the model of cooperative covering proposed by Berman et al. (IEE Trans. 40, 232–245, 2010, [1]) as an extension to classical covering problems. A key ingredient of the model is the estimation of the coverage probability by multiple facilities. We introduce a simple parametric model for the travel time of ambulances and propose two methods to calculate the coverage probability approximately. We report and discuss two solutions obtained from computations using actual data.

Detection of abrupt changes in time series data structure is very useful in modeling and prediction in many application areas, where time series pattern recognition must be implemented. Despite of the wide amount of research in this area, the proposed methods require usually a long execution time and do not provide the possibility to estimate the real changes in variance and autocorrelation at certain points. Hence they cannot be efficiently applied to the large time series where only the change points with constraints must be detected. In the framework of the present paper we provide heuristic methods based on the moving variance ratio and moving median difference for identification of change points. The methods were applied for behavior analysis of farm animals using the data sets of accelerations obtained by means of the radio frequency identification (RFID).

Reliable order promising is a key competitive factor for MTO companies. Several measures to maintain reliability in an uncertain production system are proposed in literature, but so far a multidimensional anticipation of customer response to modified order specifications has not been adequately taken into account. The purpose of this paper is to demonstrate the potential of this measure by modeling and numerically analyzing the impacts on reliability and profit.

This work focuses on the K-staged two-dimensional cutting stock problem with variable sheet size. High-quality solutions are computed by an efficient beam-search algorithm that exploits the congruency of subpatterns and takes informed decisions on which of the available sheet types to use for the solutions. We extend this algorithm by embedding it in a sequential value-correction framework that runs the algorithm multiple times while adapting element type values in each iteration and thus constitutes a guided diversification process for computing a solution. Experiments demonstrate the effectiveness of the approach and that the sequential value-correction further increases the overall quality of the constructed solutions.

We consider the 3-staged Strip Packing Problem, in which rectangular items have to be arranged onto a rectangular strip of fixed width, such that the items can be obtained by three stages of guillotine cuts while the required strip height is to be minimized. We propose a new logic-based Benders decomposition with two kinds of Benders cuts and compare it with a compact integer linear programming formulation.

Production levelling (Heijunka) is one of the key elements of the Toyota Production System and decouples customer demand from production orders. For the decoupling period a levelling pattern has to be designed. Existing approaches for the design of levelling patterns are majorly limited to large-scale production. Therefore, this article proposes a novel optimization model regarding the requirements of lot-size production. Relevant, sequence-dependent changeovers are considered. An integer, combined lot-sizing and scheduling model is formulated. The four target criteria changeover times, smoothness of daily workload, variance of lot-sizes and similarity of production sequences are aggregated into one optimization model. In a real case study of an existing production plan a clear improvement of changeover times, similarity and smoothness of workloads is realized.

Data envelopment analysis (DEA) is a method for calculating relative efficiency as a ratio of weighted outputs to weighted inputs of decision making units (DMUs). It is well-known that DEA can be done under the assumption of constant returns to scale (CRS) or variable returns to scale (VRS). One major disadvantage of the classical approach is that each DMU optimizes its individual weighting scheme–often called self-appraisal. To overcome this flaw cross-efficiency evaluation has been developed as an alternative way of efficiency evaluation and ranking of DMUs. Here all individual weighting schemes–called price systems–are applied to the activities of all DMUs. The derived cross-efficiency matrix can form the basis for seeking a consensual price system–a peer–, and hence this price system can be used for a peer-based activity planning. The present contribution shows that a scale-efficiency gap can occur when peer-based activity planning under VRS is applied, i.e. there is no feasible point in which self-appraisal efficiency under CRS, VRS and peer-appraisal efficiency under VRS coincide. As a consequence, we propose a mixed integer linear problem to avoid this drawback.

In this research we empirically verify the relationship between operational disruptions and economic cycles in the manufacturing industry of the United States. Contemporary and lagged correlation estimates are measured to demonstrate the degree of co-movement between the severity and the number of operational disruptions and several macroeconomic variables. Our findings suggest that the severity of operational disruptions follows the economic cycles with a lag of two years.

A two-level inventory system under a periodic review with lateral transshipments is considered. The supply chain is composed of the external manufacturer, the central warehouse and three identical retail outlets. By moving stock between retail outlets, the supply chain can maintain a service target while decreasing inventories. The aim is to optimize the order-up-to levels under a fill rate constraint. We combine a simulation with a barycentric interpolation at the Chebyshev points and a degree reduction technique to construct low-degree polynomial tensor product surfaces for the objective function and the constraint. The approximate optimization problem is solved by simulated annealing.

One innovative production concept is currently highly discussed in the process industry: transformable, modular plant designs implemented in standardized transportation iso-containers Buchholz (Chem Eng Process: Process Intensif, 49(10):993–995, 2010, [3]).

A game decision support tool is developed to suggest the best conditions for the coordination contract between different stakeholders with conflictive objectives in a multi-participant Supply Chain (SC). On the base of dynamic games, the interaction between the involved stakeholders is modeled as a non-cooperative non-zero-sum Stackelberg’s game under the leading role of one of the partners. The leader designs the first game move (price offered) based on its optimal conditions and taking into consideration the uncertain conditions of the follower. Consequently, the follower responds by designing the second move (quantity offered at this price) based on its best current/uncertain conditions, until the Stackelbergs payoff matrix is built. The expected follower payoffs are obtained taking into consideration the risks associated with the uncertain nature of the 3rd party suppliers. Results are verified on a case study consisting of different providers SC around a client SC in a global decentralized scenario. The results show improvements in the current/expected individual profits in the SCs of both leader and follower when compared with their standalone cases.

The “Intelligent work preparation based on virtual tooling machines” research project presents an idea for pursuing an automatically optimized machine setup to obtain minimized tool paths and production time for CNC tooling machines. A simulation based optimization method was developed and will be combination with a virtual tooling machine to validate the setup parameters and configuration scenarios. The features of the machine simulation such as material removal and collisoon detection are associated with a sharp increase in the simulation complexity level which leads to a high effort for a simple simulation based optimization approach where a high number of iterations are typically necessary to evaluate the optimization results. This contribution focuses on the implementation of a machine setup optimization in a way that is practical as pre-processing estimation for workpiece positions. Therefore a simulation using a rastered workspace model, combined with an asynchronous PSO implementation will be introduced to avoid needless simulation runs.

Within automotive supply chains, instability of order schedules of original equipment manufacturers (OEMs) creates inefficiencies in suppliers’ production processes. Due to the market power of the OEM, first tier suppliers are not always able to influence the scheduling behavior of their customers. However, addressing the root causes of schedule instability, in particular the unreliability of suppliers’ production processes, can help to curtail short-term demand variations and increase the overall supply chain efficiency. To this end, we introduce a stylised assembly supply chain model with two suppliers and a single OEM. This supply chain can be disrupted by a shortage occurring at one of the two suppliers due to random machine breakdowns, what consequently creates dependent requirements variations affecting both the buyer and the other supplier. Therefore the paper at hand contains two main sections. At first, a simulation model is developed containing the said mechanism causing schedule instability. Secondly, a simulation study is carried out to derive managerial and theoretical implications accordingly.

A three stages, linear, push-pull, production-inventory system is investigated. The system consists of a production station, a finished goods buffer, and a retailer following continuous review (s, Q) policy. Exponentially distributed production and transportation times are assumed. External demand is modeled as a compound Poisson process, and a lost sales regime is assumed. The system is modeled as a continuous time—discreet space Markov process using Matrix Analytic methods. An algorithm is developed in MatLab to construct the transition matrix that describes the system for different parameters. The resulting system of linear equations provides the vector of the stationary probabilities, and then key performance measures such as customer service levels, average inventories etc. are computed. The proposed model can be used as a descriptive model to explore the dynamics of the system via different scenarios concerning structural characteristics. Also, it may be used as an optimization tool in the context of a prescriptive model.

This paper proposes a concept for automated root cause analysis, which integrates heterogeneous data sources and works in near real-time, in order to overcome the time-delay between failure occurrence and diagnosis. Such sources are (a) vehicle data, transmitted online to a backend and (b) customer service data comprising all historical diagnosed failures of a vehicle fleet and the performed repair actions. This approach focusses on the harmonization of the different granularity of the data sources, by abstracting them in a unified representation. The vehicle behavior is recorded by raw signal aggregations. These aggregations are representing the vehicle behavior in a respective time period. At discrete moments in time these aggregations are transmitted to a backend in order to build a history of the vehicle behavior. Each workshop session is used to link the historic vehicle behavior to the customer service data. The result is a root cause database. An automatic root cause analysis can be carried out by comparing the data collected for an ego-vehicle, the vehicle the failure situation occurred, with the root cause database. On the other hand, the customer service data can be analyzed by an occurred failure code and filtered by comparing the vehicle behavior. The most valid root cause is detected by weighting the patterns described above.

Social networks are a pervasive phenomenon. While commonly exploited in industry, they are still largely unexplored from the scientific point of view, leaving a huge application potential unexpressed. Their study is hardened by two important factors: the high complexity of the systems at hand and the large amount of data to be considered. In this work we propose Integer Linear Programming (ILP) models to analyze the diffusion of knowledge through social networks. We assume a set of individuals and a set of topics to be given. Each individual has a certain level of interest and skill on each topic, that change through interactions with other individuals. Links among individuals evolve according to these interactions. As shown in the literature such a phenomenon well represents the dynamics of opinions, relationships and trust. Our ILP models are suitable for both predictive and prescriptive analytics. In particular, they can be used (a) to predict the skill level on each topic for each individual, by taking as data a sampling of the status of network links during a certain time horizon (b) to predict the status of network links, by taking as data a sampling of skill levels (c) to indicate which individuals affect most the network when their own skill is artificially increased (d) to indicate which missing links would improve the average skill level of the network. We present computational results, exploiting a simulation tool from the literature, and considering networks with up to fifty individuals, twelve topics and thousands of time steps. These show that out ILP approach is computationally viable also on large scale data, requires very few parameters to be tuned during training, and provides results of reasonable accuracy, especially in tasks (a) and (c).

Computational Topological Data Analysis (TDA) is a collection of procedures which permits extracting certain robust features of high dimensional data, even when the number of data points is relatively small. Classical statistical data analysis is not very successful at or even cannot handle such situations altogether. Hidden features or structure in high dimensional data expresses some direct and indirect links between data points. Such may be the case when there are no explicit links between persons like clients in a database but there may still be important implicit links which characterize client populations and which also make different such populations more comparable. We explore the potential usefulness of applying TDA to different versions of credit scoring data, where clients are credit takers with a known defaulting behavior.

This paper introduces a concept for product optimization support based on the integration of customer data sources. The motivation is a common misunderstanding gap between the manufacturer and the customer. While the customer has certain needs, the manufacturer aims at embedding them into the product design. However, due to imprecise understanding of the needs and subsequent development mistakes, the product can vary from what the customer actually requires. The concept combines two different data sources in order to reveal the gap between the product and the customer needs. The first source is represented by a customer-product interaction log file. The second source is social media delivering customer feedback regarding the product.

This paper discusses the validity of the frequent shoppers program (FSP) based on a particle filter estimation model.

We apply wavelet decomposition to decompose financial return series into a time frequency domain and assess the relevant frequencies for adequate daily Value-at-Risk (VaR) forecasts. Our results indicate that the frequencies that describe the short-run information of the underlying time series comprise the necessary information for daily VaR forecasts.

The combination of governmental incentives and falling module prices has led to a rapid increase of globally installed solar photovoltaic (PV) capacity. Consequently, solar power becomes more and more important for the electricity system. One main challenge is the volatility of solar irradiance and variable renewable energy sources in general. In this context, accurate and reliable forecasts of power generation are required for both electricity trading and grid operation. This study builds and evaluates models for day-ahead forecasting of PV electricity feed-in. Different state-of-the-art forecasting models are implemented and applied to a portfolio of ten PV systems. More specifically, a linear model and an autoregressive model with exogenous input are used. Both models include inputs from numerical weather prediction and are combined with a statistical clear sky model using the method of weighted quantile regression. Forecasting-related economic implications are analyzed by means of a two-dimensional mean-variance approach. It is shown that enhanced forecast accuracy does not necessarily imply an economic gain.

In this paper we analyze the impact of heterogeneous agents and learning effects on negotiated transfer prices and the consolidated profit resulting at firm level. An agent-based simulation is employed to show potential results implied by learning and interaction effects between negotiating profit centers. In particular, intra-company profit centers can choose to trade with each other or with independent parties on an external market. Since the profit centers have incomplete and heterogeneous information about this external market, they are involved in a bargaining process with outside options. To achieve a maximized comprehensive income it may be favourable on profit center level or even on firm level to choose outside options. In the long run the intracompany option should be favourable on all levels, as it excludes the profit orientated external market. We investigate our agents’ behaviour under different parameter settings regarding the incentive system set by the company-wide management. Results show how learning effects and different incentive systems affect the decision making process with respect to the firm’s overall objective.

Two coupling schemes where probabilities of credit rating migrations vary across industry sectors are introduced. Favorable and adverse macroeconomic factors, encoded as values 1 and 0, of credit class- and industry-specific unobserved tendency variables, modify the transition probabilities rendering individual evolutions dependent. Unlike in the known coupling schemes, expansion in some industry sectors and credit classes coexists with shrinkage in the rest. The schemes are tested on Standard and Poor’s data. Maximum likelihood estimators and MATLAB optimization software were used.

Currently, the major challenge in the life insurance sector is to find a numerically efficient and precise method for the estimation of the fair value of future liability cash flows. Besides least square Monte Carlo algorithms, the construction of replicating portfolios is very popular. However, there has been a debate as to how diversions between future discounted cash flows of the replicating portfolio and liabilities ought to be penalized. A frequently used argument against squared error penalization is that a few scenarios with abnormally high interest rates will cause big discrepancies between future cash flows. These scenarios will therefore dominate in the minimization with the consequence that the replicating portfolio badly approximates liabilities in the average scenario. In this article we undermine this argument by showing that the described observation will not take place when discounting with the appropriate numéraire.

We propose a model where two competitors, a Leader and a Follower, are sequentially creating their hub and spoke networks and setting prices. The existence of the unique Stackelberg and Nash pricing equilibria is shown. On the basis of these results we give the conclusion about existence of the profit maximising solution for the Leader.

A numerical procedure for adaptive parameterization of changing 3D objects for knowledge representation, analysis and optimization is developed. The object is not a full CAD model since it involves many shape parameters and excessive details. Instead, optical 3D scanning of the actual object is used (stereo-photogrammetry, triangulation) which leads to the big-data territory with point clouds of size $$10^8$$ and beyond. The total number of inherent surface parameters corresponds to the dimensionality of the shape optimization space. Parameterization must be highly compact and efficient while capable of representing sufficiently generic 3D shapes. The procedure must handle dynamically changing shapes in optimization quasi-time iterations. It must be flexible and autonomously adaptable as edges and peaks may disappear and new ones may arise. Adaptive re-allocation of the control points is based on feature recognition procedures (edges, peaks) operating on eigenvalue ratios and slope/ curvature estimators. The procedure involves identification of areas with significant change in geometry and formation of partitions.

The distortionary effect of upstream petroleum taxation has been discussed extensively by economists. The literature however, has largly neglected the Production Sharing Contract (PSC) which is widely used by the internationl petroleum industry. We examine how a PSC can distort the optimal time path of production from an oil reservoir. To do that, we use optimal control theory and solve the problem with Hamiltonian function. We show that, regardless of the contract parameters, a PSC always distort the time path of production unless the oil price changes at the rate of interest rate.

In this paper we present an application of dynamic tracking games to a monetary union. We use a small stylized nonlinear two-country macroeconomic model of a monetary union for analysing the interactions between two fiscal (governments) and one monetary (common central bank) policy makers. We introduce a negative asymmetric demand side shock describing the macroeconomic dynamics within a monetary union similar to the economic crisis (2007–2010) and the sovereign debt crisis (since 2010) in Europe. We investigate the welfare consequences of three scenarios: fiscal policies by independent governments (the present situation), centralized fiscal policy (a fiscal union) with an independent central bank, and a fully centralized fiscal and monetary union. For the latter two scenarios, we investigate the effects of different assumptions about the weights for the two governments in the cooperative agreement.

In air traffic management, most optimization procedures are commonly based on deterministic modeling and do not take into account the uncertainties on environmental conditions (e.g., wind) and on air traffic control operations. However, aircraft performances in a real-world context are highly sensitive to these uncertainties. The aim of this work is twofold. First, we provide some numerical evidence of the sensitivity of fuel consumption and flight duration with respect to random fluctuations of the wind and the air traffic control operations. Second, we develop a global stochastic optimization procedure for general aircraft performance criteria. Since we consider general (black-box) cost functions, we develop a derivative-free optimization procedure: noisy simulated annealing (NSA).

Different reasons can hinder the application of multiparametric programming formulations to solve optimization problems under uncertainty, as the high nonlinearity of the optimization model, and/or its complicated structure. This work presents a complementary method that can assist in such situations. The proposed tool uses kriging metamodels to provide global multiparametric metamodels that approximate the optimal solutions as functions of the problem uncertain parameters. The method has been tested with two benchmark problems of different characteristics, and applied to a case study. The results show the high accuracy of the methodology to predict the multiparametric behavior of the optimal solution, high robustness to deal with different problem types using small number of data, and significant reduction in the solution procedure complexity in comparison with classical multiparametric programming approaches.

In primary care mainly two types of patient requests occur: walk-ins without an appointment and patients with a prescheduled appointment. The number and position of such prescheduled appointments influence waiting times for patients, capacity for treatment and the utilization of physicians. An integer linear model is developed, where the minimum number of appointments prescheduled for a weekly profile is determined. Since the number of patient requests differs significantly between seasons, weekdays and daytime, efficient appointment scheduling has to take different scenarios into account. Using an intensive monte-carlo simulation, we compare appointment strategies with respect to their performance for different scenarios.

For patients requesting emergency medical services (EMS) in a life-threatening emergency, the probability of survival is strongly related to the rapidness of assistance. An especially challenging task for planners is to allocate limited resources while managing increasing demand for services. The provision of sufficient staff resources for the ambulances has great impact on the initial treatment of patients and thus on the quality of emergency services. Data-driven empirically required ambulance location planning as well as the allocation of staff for these vehicles are successively optimized in the proposed approach to support emergency medical service decision makers. According to the identified problem structure, an integer linear programming model is proposed. An exemplary case study based on real-world data demonstrates how this approach can be used within the emergency medical service planning process.

We present a fairly involved ILP-model for a complex personnel planning problem arising from a real-world application. Two main aspects distinguish the problem from standard models: (i) Several tasks may be executed by the same person simultaneously. However, this multi-tasking is subject to certain complicated conditions. (ii) Qualification of personnel is complex and totally inhomogeneous. We introduce a representation for both issues that is at the same time fairly general and still easy enough to operate for the personnel manager.

Energy-efficient components do not automatically lead to energy-efficient systems. Technical Operations Research (TOR) shifts the focus from the single component to the system as a whole and finds its optimal topology and operating strategy simultaneously. In previous works, we provided a preselected construction kit of suitable components for the algorithm. This approach may give rise to a combinatorial explosion if the preselection cannot be cut down to a reasonable number by human intuition. To reduce the number of discrete decisions, we integrate laws derived from similarity theory into the optimization model. Since the physical characteristics of a production series are similar, it can be described by affinity and scaling laws. Making use of these laws, our construction kit can be modeled more efficiently: Instead of a preselection of components, it now encompasses whole model ranges. This allows us to significantly increase the number of possible set-ups in our model. In this paper, we present how to embed this new formulation into a mixed-integer program and assess the run time via benchmarks. We present our approach on the example of a ventilation system design problem.

German Railways Infrastructure division DB Netz has started to gradually introduce a new process for its rail freight timetabling. This process contains two main stages: at first a pre-planning of standardized train paths (called slots) and afterwards the assignment of train path applications to the pre-planned slots. The current implemented train path assignment optimization model by Nachtigall and Opitz 2014 has a model scope of one single traffic day. However, current train path applications for the network timetable have multiple and diverse traffic days (e.g. Monday till Friday or Tuesday and Thursday) and cannot be assigned appropriately today. At first it will be illustrated by what a good train path assignment with multiple traffic days is characterized from the point of view of an infrastructure manager and its customers. Subsequently, a new hierarchical approach for train path assignment with different traffic days will be presented which is derived from the analysis of present train path applications. The analyses of a German long term timetable scenario indicate that this new approach generates promising results.

The German Railway Company (DB Netz) schedules freight trains by connecting pre-constructed slots to a full train path. We consider this problem with special attention to traffic day restrictions and model it by a binary linear decision model. For each train request a train path has to be constructed from a set of pre-defined path parts within a time-space network. Those train requests should be realized only at certain days of the week. Each customer request has a specific traffic day pattern, which is a difficult challenge for the allocation process. Infrastructure capacity managers intend to achieve an efficient utilization of the capacity, whereas customers are interested in homogeneous train paths, i.e. they want the same traffic path connection for all requested traffic days. We discuss those partly contradictory requirements within the context of our binary linear decision model. The problem is solved by using column generation within a branch and price approach. We give some modeling and implementation details and present computational results from real world instances.

Hydro storage system optimization is becoming one of the most challenging task in Energy Finance. Following the Blomvall and Lindberg (2002) interior point model, we set up a stochastic multiperiod optimization procedure by means of a “bushy” recombining tree that provides fast computational results. Inequality constraints are packed into the objective function by the logarithmic barrier approach and the utility function is approximated by its second order Taylor polynomial. The optimal solution for the original problem is obtained as a diagonal sequence where the first diagonal dimension is the parameter controlling the logarithmic penalty and the second is the parameter for the Newton step in the construction of the approximated solution. Optimimal intraday electricity trading and water values for hydroassets are computed. The algorithm is implemented in Mathematica.

The promotion of electricity from renewable energy in Germany by means of guaranteed feed-in tariffs and preferential dispatch leads to difficulties in the profitable operation of many modern conventional power plants. Nevertheless, conventional power generation technologies with enhanced flexibility in their operational characteristics can contribute to balancing electricity supply and demand. For this reason, the operational flexibility of conventional power plants becomes important for the system and has an inherent economic value. The focus of this research is on high efficiency gas-fired power plants; we tackle the following three research questions from a plant owner’s perspective: (1) How can already existing conventional power plants be operated more flexibly and thus be made more profitable? (2) Which flexibility measures can be taken under consideration? (3) What is the optimal timing to invest in flexibility measures? To answer these questions we propose an optimization model for the flexible operation of existing gas-fired power plants that is based on real options analysis (ROA). In the model, the economic and technical aspects of the power plant operation are explicitly taken into account. Moreover, the spark spread, which is an important source of uncertainty, is used for the definition of the flexible plant operation in terms of different load levels and corresponding efficiency factors. The usefulness of the proposed model is illustrated with a case study mimicking the retrofitting decision process.

Since the liberalisation of the German energy markets, supply companies strive to gain additional revenues by trading in different electricity markets. Thus, trading strategies have to be adjusted to lower but more volatile energy prices caused by a rising share of renewable energy feed-in. Due to the increasing importance of combined heat and power (CHP) plants, an energy supply company is considered that operates a CHP plant with heat storage. A new modelling approach is presented to support generation companies scheduling their participation in two sequential electricity markets, the balancing market and the day-ahead spot market. In both markets, specific bidding procedures for trading exist. To derive optimal bidding strategies for CHP plants, we formulate the bidding problem as an innovative and detailed multistage stochastic programming model taking into account the sequencing of market clearing. The optimisation model simultaneously determines the operation of the CHP plant and heat storage device. An exemplary case study illustrates the benefit from coordinated bidding in sequential electricity markets.

In this pit drainage study, we investigate how pump control optimized with respect to the prevailing electricity prices impacts the operating costs. The optimization of the well operation takes the dependency of the electrical power on the water volume lifted and the changing water levels into account. The nonlinear dependency is transformed into a linear optimization problem in multiple stages. First, a superstructure optimization is used. Second, the characteristic pump profiles are linearized piecewise, resulting in a simplified problem where only the multiplication of a binary and a positive real variable remains. The multiplication of the two variables is replaced by a new variable, transforming the optimization problem into a mixed integer linear optimization (MILP) problem. The results of the superstructure optimization yield the optimal pump size and the minimal costs incurred, which are then used to optimize the maintenance strategy. We find that well costs can be reduced markedly by the optimization proposed.

This paper investigates the future need of the Brazilian electric grid for High-Voltage-Direct-Current (HVDC) transmission lines using expansion scenarios. Currently, electricity is produced mainly in large hydropower plants with enormous reservoirs, but the model is approaching a limit because the potential for hydropower near load centers is almost depleted. To preserve a low-carbon electrical energy mix, renewable energy sources in far more distant locations need to be exploited. This paper focuses on the expansion potential of wind, solar and hydropower and its spatial mismatch with the expected future electricity demand. Linear optimization is used to determine the best HVDC connections. The results show clear differences in the costs for transmission lines in different scenarios, confirming that the transmission capacity is a critical factor in the process of expanding the renewable energy generation capacity in Brazil. This study provides insights for policy makers and industry.

In the last decade the share of renewable feed-in increased sharply. More than 25 % of the gross electricity consumption in Germany is typically covered by renewable sources. Despite the positive effects, e.g., low marginal costs and a sustainable electricity supply with less emissions, a generation portfolio should also contain thermal power plants as well as energy storages, mainly hydro storages, to guarantee a long-term security of electricity supply. The problem of flexible coordination and dispatch of thermal power plants and hydro storages in order to meet the highly volatile residual demand (energy demand minus the volatile feed-in of renewables) becomes even more challenging for generating companies. Therefore, we present a new two-stage heuristic approach for solving the resulting long-term unit commitment problem with hydro-thermal coordination (UCP-HT), where system operating costs have to be minimized. Within a comprehensive performance analysis, the results of the approach are compared to near-optimal solutions obtained by CPLEX on the basis of a tight mixed-integer linear formulation for the UCP-HT. In particular, well-known instances from the literature with real-world energy demands, a planning horizon of one year, and hourly time steps are solved within minutes ensuring a solution gap of less than 1 %.

The advent of smart electricity meters is bound to affect the electricity distribution sector in more ways than one can imagine. Thus, companies involved in the retail electricity market will need to adjust to the new conditions, especially with regard to the electricity pricing, which constitutes an important part of the activities of a retailer. In this work, a bilevel model is proposed as a tool for facilitating the retailer’s task in optimally selecting price levels to be announced to various local resources composing his portfolio. Application of proper linearizations allows obtaining and equivalent one-level mixed integer linear programming problem (MILP) which is solved using CPLEX solver under GAMS. Examination of several scenarios regarding the form of the pricing scheme shows that the number of time zones as well as the grouping/assignment of each hour of the day to one time zone or another affects significantly the profitability of the retailer’s business.

The market for Energy Performance Contracting (EPC) in Russia is just emerging, but progress has so far been slow despite of promising forecasts. The successful realization of EPC projects requires a sound understanding of the main project risks. It highly depends on effective risk analysis and management, which should be essential parts of daily business activities of Energy Service Companies (ESCOs) and Energy Service Providing Companies (ESPCs) engaged in EPC projects in Russia. This study provides a first risk analysis framework for EPC projects executed in the industrial sector in Russia. General risks associated with EPC projects identified from the international literature were validated by Russian EPC practitioners in expert interviews. An Analytic Hierarchy Process approach was used to rank the identified risk factors and causes of risk in terms of their contribution to the riskiness of EPC projects. The data were obtained from a web-based survey among the Russian ESCOs and ESPCs. A widely usable risk management framework is proposed that is potentially very useful for EPC practitioners and can support the development of proactive risk analysis in EPC projects required both by third-party capital lenders and the further development of EPC market regulation. We also identify causes of risk related to the financial and regulatory aspects contributing the most to the riskiness of industrial sector EPC projects in Russia.

Distributed energy resource (DER) systems, composed of different small to medium-sized renewable and conventional energy generators, allow to balance the volatile supply from renewable energies. Considering a DER system with power and district heat, small combined heat and power (CHP) plants and their flexible operation play a central role. Flexibility investments such as heat storage devices provide further flexibility for the DER system. In this contribution, the profitability of an investment in a heat storage device and its optimal capacity are examined. An innovative decision support approach for capacity determination of such flexibility investments is applied considering the investment’s benefit throughout its entire lifetime. System dependency as well as the decision maker’s risk attitude are taken into account. The developed two-stage stochastic programming model simultaneously optimizes the operation of the DER system and the investment’s capacity. In a simulation study, the calculated capacities are evaluated. An exemplary case study illustrates the advantages of the proposed approach.

The evaluation and selection of intermodal routes with regard to the key objectives, i.e., transit time, transport emissions and cost, is the main challenge in the design of intermodal networks. The aim of this paper is to present a decision support system for intermodal freight transportation planning, which offers methodological contributions to the research on transport mode, route and carrier selection as well as results for industrial practitioners for the assessment of emission abatement potentials. Core of this approach is a capacitated multi-commodity network flow model considering three minimization objectives, i.e. costs, time and CO$$_2$$-equivalents. In this contribution a tri-objective mixed-integer linear model formulation minimizes the number of transported and transshipped full truck loads taking into account tied in-transit capital and the distance travelled. The decision support system is validated in an exemplary case study application analyzing the sensitivity of objectives on optimal route and carrier choice. By applying the augmented $$\varepsilon $$-constraint method, a Pareto-efficient frontier is determined to investigate the tradeoff between economic and ecological objectives in intermodal freight transportation planning.

The complexity of large-scale power system models often necessitates the choice of a suitable temporal resolution. Nowadays, mainly simple heuristic approaches are used. An adequate decision support related to power generation and transmission optimisation in systems with a high RES share, however, requires preserving the complex intra-period and intra-regional links within and between the volatile electricity demand and supply profiles. Focussing on power systems operation, we are able to show that even an amount of less than 300 time segments may be sufficient for the modelling of a whole year, if chosen carefully.

In this paper the worldwide gas market model WEGA and the base case scenario of Stadtwerke München (SWM) are described. The potential of WEGA in practical application is shown through three sensitivity analyses. Firstly, it is demonstrated that shale gas in Europe has only a small impact on European gas prices in future. Nevertheless, shale gas is very important for security of supply in Europe. The second sensitivity analysis is calculated with modified U.S. liquefied natural gas (LNG) export volumes. A surge of LNG would significantly decrease the gas price in Europe. On the other hand, Europe will be oversupplied till the end of the twenties and consequently, the prices will be relatively immune to few U.S. LNG export volumes in this period. At the third sensitivity analysis, gas demand is modified. The results reveal that the gas prices are very sensitive to changes of demand.