Search

In business process landscapes, a common challenge is to provide the necessary computational resources to enact the single process steps. One well-known approach to solve this issue in a cost-efficient way is to use the notion of elasticity, i.e., to provide cloud-based computational resources in a rapid fashion and to enact the single process steps on these resources. Existing approaches to provide elastic processes are mostly based on Virtual Machines (VMs). Utilizing container technologies could enable a more fine-grained allocation of process steps to computational resources, leading to a better resource utilization and improved cost efficiency. In this paper, we propose an approach to optimize resource allocation for elastic processes by applying a four-fold auto-scaling approach. The main goal is to minimize the cost of process enactments by using containers. To this end, we formulate and implement a multi-objective optimization problem applying Mixed-Integer Linear Programming and use a transformation step to allocate software services to containers. We thoroughly evaluate the optimization problem and show that it can lead to significant cost savings while maintaining Service Lev

Business Process Management Systems (BPMS) need to be able to take into account the fluctuating demand for computational resources during the execution of business process activities. Today, BPMS rely on the leasing and releasing of virtual machines (VMs) on cloud resources, which leads to a rather coarse-grained allocation of computational resources. This may result in an increase in the execution cost, flexibility restrictions, and a negative impact on the Quality of Service. In order to overcome these drawbacks, we introduce the Vienna Platform for Elastic Processes on Containers (ViePEP-C). ViePEP-C is an elastic BPMS that uses containers instead of VMs for the execution of business process activities on cloud resources, leading to a more fine-grained execution environment. To achieve this, ViePEP-C offers cloud controller, monitoring and business process execution functionalities and provides a platform for different resource and task scheduling algorithms. To evaluate the benefits of ViePEP-C, we further present a resource and task scheduling algorithm and show that, by using containers as execution environment, the execution cost can be decreased by over 20% (compared to a state-of-the-art VM-based scheduling algorithm) while considering a high service level.

Process mining supports business process management with operational insights extracted from event logs. A key challenge for process mining is that operational processes in production and logistics often include batching and unbatching, e.g., to delivery several packages using one truck tour. Such n:m relations blur the notion of a process instance and make the causality between events difficult to trace. In this paper, we address this research problem by introducing causal event models that capture batching behavior accurately. To this end, we construct conflict-free prime event structures for event instances of the event log, and devise various analysis techniques on top of them. We implemented the techniques in a tool and run in real data of a manufacturing company with various 1:n and n:1 relations in their production process showing the potential of our approach.

(a) Situation faced: The accounting function of the Deutsche Bahn Group, a public German railway company with 320,000 employees, went through a centralization process to transfer the entire company’s accounting activities from 350 locations to three Shared Service Centers. The decentralized accounting function prior to this transformation was qualitatively and economically unsustainable. (b) Action taken: In the course of the transformation, processes had to be changed on a global level, and a workflow management system had to be introduced to automate these processes. The TIM BPM Suite, a workflow management system, was selected among several digital tools as the main pillar to make this transition possible. The organizational desire for further automation led to efforts to combine the workflow with existing third-party systems to meet regulatory requirements, to achieve faster turnaround times, and to introduce performance monitoring on a global scale. These goals were realized by using robotic process automation in combination with human workflows. (c) Results achieved: The newly centralized accounting function guarantees globally standardized digital processes. Thanks to its flexible rollout options, positive economies of scale were achieved that facilitate the integration of existing companies and new acquisitions. Moreover, the new approach meets all legal and compliance requirements. The project won the GFO organization’s “Process Solution Award 2019” in the category of workflow management. (d) Lessons learned: Acceptance of the new way to work is one of the most important key aspects of such a transformation, especially when the transformation changes the way the employees work. The early integration of all employees is essential to ensure quick acceptance.

The use of cloud-based storage systems for storing data is a popular alternative to local storage systems. Beside several benefits of cloud-based storages, there are also downsides like vendor lock-in or unavailability. Moreover, the selection of the best fitting storage solution can be a tedious and cumbersome task and the storage requirements may change over time. In this paper, we formulate a system model that uses multiple cloud-based services to realize a redundant and cost-efficient storage. Within this system model, we formulate a local and a global optimization problem that considers historical data access information and predefined quality of service requirements to select a cost-efficient storage solution. Furthermore, we present a heuristic optimization approach for the global optimization. Extensive evaluations show the benefits of our work in comparison with a baseline that follows a state-of-the-art approach. We show that our solutions save up to 30% of the cumulative cost in comparison with the baseline.

A new requirement for the manufacturing companies in Industry 4.0 is to be flexible with respect to changes in demands, requiring them to react rapidly and efficiently on the production capacities. Together with the trend to use Service-Oriented Architectures (SOA), this requirement induces a need for agile collaboration among supply chain partners, but also between different divisions or branches of the same company. In order to address this collaboration challenge, we~propose a novel pragmatic approach for the process analysis, implementation and execution. This~is achieved through sets of semantic annotations of business process models encoded into BPMN 2.0 extensions. Building blocks for such manufacturing processes are the individual available services, which are also semantically annotated according to the Everything-as-a-Service (XaaS) principles and stored into a common marketplace. The optimization of such manufacturing processes combines pattern-based semantic composition of services with their non-functional aspects. This is achieved by means of Quality-of-Service (QoS)-based Constraint Optimization Problem (COP) solving, resulting in an automatic implementation of service-based manufacturing processes. The produced solution is mapped back to the BPMN 2.0 standard formalism by means of the introduced extension elements, fully detailing the enactable optimal process service plan produced. This approach allows enacting a process instance, using just-in-time service leasing, allocation of resources and dynamic replanning in the case of failures. This proposition provides the best compromise between external visibility, control and flexibility. In this way, it provides an optimal approach for business process models' implementation, with a full service-oriented taste, by implementing user-defined QoS metrics, just-in-time execution and basic dynamic repairing capabilities. This paper presents the described approach and the technical architecture and depicts one initial industrial application in the manufacturing domain of aluminum forging for bicycle hull body forming, where the advantages stemming from the main capabilities of this approach are sketched.

Cloud manufacturing supports companies to create cross-organizational elastic process landscapes with flexible and scalable processes. In recent years, the Business Process Model and Notation language gained interest in the cloud manufacturing domain not only as a modeling language but also as a base for process enactment. During this enactment of manufacturing processes, accurate knowledge about when a machine is needed is of great importance. While such planning is simple in a static production environment, it becomes a challenging task in an elastic cloud manufacturing environment with a variety of flexible and scalable processes. Hence, we propose an approach that performs enactment simulations before an actual manufacturing process is carried out. The goal is to reserve manufacturing machines at the time at which they will be needed in a machine reservation timetable. By discussing different use case scenarios, we further elaborate on the benefits of our approach.

A key challenge for elastic business processes is the resource-efficient scheduling of cloud resources in such a way that Quality-of-Service levels are met. So far, this has been difficult, since existing approaches use a coarse-granular resource allocation based on virtual machines.

One of the most important requirements for the manufacturing industry is the scalable and fault tolerant realization of their business processes. While there are already several propositions towards elastic process execution on cloud resources, compensation for faulty tasks often remains a manual task. In this paper, we present FCE4BPMN, which realizes a cloud-based execution environment, which can compensate business model executions exceptions on-demand. These compensations are performed just-in-time and ad-hoc, based on a semantically annotated BPMN model. Besides basic compensations, the optimization component is also capable of optimizing the overall process execution based on different QoS criteria. We, therefore, provide an extension to the BPMN standard as well as an execution environment to run business processes on cloud resources. Finally, we validate our approach based on a set of requirements originating from the manufacturing domain.

The Internet of Things is getting more and more traction, nevertheless, state-of-the-art approaches only focus on specific aspects, like the integration of heterogeneous devices or the processing of sensor data emitted by these devices. However, such domain-specific approaches slow the adoption rate of the Internet of Things, because users need to select and integrate different approaches in order to build a solution that fits all their requirements. To resolve this shortcoming, we have designed and implemented the VISP ecosystem, which provides a holistic approach for elastic data stream processing in Internet of Things scenarios by supporting the complete lifecycle of designing, deploying, and executing such scenarios. VISP further tackles the challenges of data privacy as well as software reuse, including monetization aspects in today's service landscapes. This paper analyzes challenges for creating solutions for the Internet of Things, presents the VISP ecosystem, and discusses its applicability for use case specific data stream processing topologies.

Nowadays, the usage of cloud storages to store data is a popular alternative to traditional local storage systems. However, besides the benefits such services can offer, there are also some downsides like vendor lock-in or unavailability. Furthermore, the large number of available providers and their different pricing models can turn the search for the best fitting provider into a tedious and cumbersome task. Furthermore, the optimal selection of a provider may change over time.In this paper, we formalize a system model that uses several cloud storages to offer a redundant storage for data. The according optimization problem considers historic data access patterns and predefined Quality of Service requirements for the selection of the best-fitting storages. Through extensive evaluations we show the benefits of our work and compare the novel approach against a baseline which follows a state-of-the-art approach.

We compare the performance and optical quality of two types of adaptive fluidic silicone-membrane lenses. The membranes feature either a homogeneous thickness, or they are shaped resulting in an inhomogeneous cross-section. The lens systems incorporate a piezoelectric actuator which is operated in a regime of ±40 V. The shaped membrane lenses show lower wave front errors than the planar ones, down to 24 nm. The pumping actuator indicates a maximum pump volume of 1159 nl at an inner piezo radius of 2.75 mm for the smallest supporting ring configuration. The full system with a planar membrane achieves a larger refractive power ranging from +19 to -14 dpt. It also shows a shorter full scale response time ( tplanar = 23.9 ms) compared to the shaped membrane ( tshaped = 35.4 ms).

Complex, message-based service systems discourage central execution control, require extremely loose coupling, have to cope with unpredictable availability of individual (composite) services, and may experience a dynamically changing number of service instances. At the topmost level, the architecture of such a complex system often follows a messaging style most naturally. A major problem during the design of these systems is achieving an overall consistent configuration (i.e, ensuring intended message routing across producers, consumers, and brokers). While orchestration or choreography-based approaches support the design of individual composite services along a workflow-centric paradigm, they are an awkward fit for specifying a message-centric architecture. In this paper, we present an architecture-centric approach to designing complex service systems. Specifically we propose modeling the system’s high-level architecture with an architecture description language (ADL). The ADL captures the message-centric configuration which subsequently allows for consistency checking. An architecture-to-configuration transformation ensures that the individual deployed services follow the architecture without having to rely on a central coordinator at runtime. Utilizing our provided tool support, we demonstrate the successful application of our methodology on a real world service system.

Tunable multi-chamber microfluidic membrane microlenses with achromaticity over a given focal length range are demonstrated. In analogy to a fixed-focus achromatic doublet lens, the multi-lens system is based on a stack of microfluidic cavities filled with optically optimized liquids with precisely defined refractive index and Abbe number, and these are independently pneumatically actuated. The membranes separating the cavities form the refractive optical surfaces, and the curvatures as a function of pressure are calculated using a mechanical model for deformation of flexible plates. The results are combined with optical ray tracing simulations of the multi-lens system to yield chromatic aberration behavior, which is verified experimentally. A focal length tuning range of 5-40 mm and reduction in chromatic aberration of over 30% is demonstrated, limited by the availability of optical fluids.

We present measurements and simulations of membrane-based micro-lens stacks, tunable in focal length in the range of 10 mm to 50 mm without chromatic aberration. The pressure-actuated, liquid-filled, membrane-based micro-lenses are fabricated by an all-silicone molding approach and consist of three chambers separated by two highly flexible silicone-membranes. Based on the idea of the classical achromatic Fraunhofer doublet, two different liquids with suitable optical properties are used. Pressure-dependent surface topologies are measured by profilometry for determining the correlation between refraction and applied pressure. The profiles are fit to polynomials; the coefficients of the polynomials are pressure-dependent and fit to empirically determined functions which are then used as an input for optical ray-tracing. Using this approach, the focal length is tunable while compensating for chromatic aberration by suitably applied pressures.

A design approach and new manufacturing technique for a novel type of stacked fluidic multi-chamber tunable lenses is presented.

We compare the performance and optical quality of two types of adaptive fluidic silicone-membrane lenses. The membranes feature either a homogeneous thickness, or it is shaped resulting in an inhomogeneous cross-section. The lens systems incorporate a piezo actuator which is operated in a regime of ±40 V. The shaped membrane lenses show lower wave front errors than the planar ones, down to 24 nm. However, the system with a planar membrane achieves a larger refractive power ranging from +19 to −14 dpt. It also shows a shorter full scale response time (t planar = 23.9 ms) compared to the shaped membrane (t shaped = 35.4 ms).

Simulation and measurements of achromatic variable-focus multi-chamber micro-lenses are presented. By using three liquid-filled chambers with well-defined dispersion behavior, separated by highly elastic silicone membranes, this achromatic optical system can be tuned in focal length by adjusting the pressures inside the chambers. Tunable achromatic micro-lenses are thus feasible.

Polymer swelling is studied as a novel method for bandgap tuning of photonic crystals (PCs). Crosslinked polymers swell in the presence of solvents, thereby increasing their volume and changing their refractive index. Consequently, the lattice constant and the refractive index of a polymer PC is modified by swelling and leads to a shift of the optical bandgap.We present experimental studies of swelling of one-dimensional PCs and theoretical calculations of swelling of three-dimensional PCs. It is shown that the frequencies of the optical bandgap of a polymer PC is shifted by about 10% when exposed to a solvent-nitrogen gas mixture at 80% of the solvent saturation pressure.

We present the dispersion characteristics of 18 liquids and one resin, which are widely used as media for liquid lenses in adaptive and tunable optics and for index matching in spectrochemical analysis. These are measured by using a refractometer operating at six different wavelengths. We provide a short description of the measurement setup and present a detailed uncertainty analysis of the measurement system to provide a measure of the reliability of the data. We conclude with a catalog of refractive indices and Sellmeier coefficients of the measured liquids and show the location of the analyzed materials in an Abbe diagram.

We present an extended optical characterization of an adaptive microfluidic silicone-membrane lens at a wavelength of 633 nm, respectively 660 nm. Two different membrane variations; one with a homogeneous membrane thickness, and one with a shaped cross section, have been realized. This paper includes the theoretical predictions of the optical performance via FEM simulation and ray tracing, and a subsequent orientation dependent experimental analysis of the lens quality which is measured with an MTF setup and a Mach-Zehnder interferometer. The influence of the fabrication process on the optical performance is also characterized by the membrane deformation in the non-deflected state. The lens with the homogeneous membrane of 5 mm in diameter and an aperture of 2.5 mm indicates an almost orientation independent image quality of 117 linepairs/mm at a contrast of 50%. The shaped membrane lenses show a minimum wave front error of WFE(RMS) = 24 nm, and the lenses with a planar membrane of WFE(RMS) = 31 nm at an aperture of 2.125 mm.