Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'How does the wind blow behind wind turbines and in wind farms?' taking place on Tuesday, 11 March 2014 at 16:30-18:00. The meet-the-authors will take place in the poster area.

Dirk Wagner FIR e.V. at RWTH Aachen University, Germany
Co-authors:
Dirk Wagner (1) F P
(1) FIR e.V. at RWTH Aachen University, Aachen, Germany

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Mr. Wagner has been working in many wind industry projects. Since 2010 he worked for a large german energy suplier. Since 2011 he is scientific research assistant and project manager at the FIR institute at RWTH Aachen University. He studied industrial Engineering (specification: mechanical Engineering) at RWTH Aachen University. He focused on operation and maintenance planning for wind turbines and worked together with (On- and Offshore) wind farm O
operators in consultancy and research projects.

Abstract

SIMULATION BASED EVALUATION OF AVAILABILITY GUARANTEES IN WIND INDUSTRY SECTOR

Introduction

The growth of installed wind capacities {1}{2} generated a markt with several service offers for operation & maintenance of wind turbines. Different parties like manufacturers, component suppliers as well as independent service providers compete for the attractive after sales market. An innovative service offer which seems to meet the customers’ requirements is the guarantee of availability {3}{4}{5}{6}. However these service providers are facing new challenges regarding their performance potentials and their financial risks occurring from possible penalties {7}. To be able to quantify these challenges and risks a simulation model has been designed in the context of a German research project named “WinServ”.

Approach

The German research project “WinServ” develops a simulation model to quantify the organizational and financial changes occurring from an innovative service offer called “availability guarantee” for wind turbines. Today many service providers offer services which are singularly requested in case of customers demand {8}. These (functional based) business models {9} imply low financial risk for the service provider and enable him to generate a demand based resource planning for his service resources. However these general conditions are changing when service providers switch their service offer into a guarantee of availability including a defined service level. In this case providers have to ensure a high flexibility regarding their resource management and build up larger capacities to be able to react quickly on unexpected turbine breakdowns. Beside the expected changes in the required resource capacities, the financial risk for service providers will also increase. In cases of high unexpected demands (e.g. after strong storms) the provider is facing several simultaneous service orders. In this situation he has to ensure a quick reaction from his service technicians otherwise he has to pay for costs occurring from long shutdown times which are covered by the agreed service level (e.g. 97% technical availability). Hence the service provider is facing the trade-off between high cost caused by his adequate resource capacities and high cost caused by penalties for neglecting his promised service level. This trade-off situation is principally already described by several authors {10}{11}{12}{13}{14}. But the described problem becomes even more relevant for service providers caused by the “availability guarantee” service offer due to the additional financial risk.

Main body of abstract

Through literature research and through workshops and interviews with the practical experts of “WinServ” research project some changing areas could be identified. The most important areas which service providers have to verify if they want to offer availability guarantees are the qualitative and quantitative resource capacities, the way how they allocate their service orders, the sharing of risks, the qualification and cooperation level of their companies and the local allocation of the service infrastructures. To find out the essential modifications in these areas a simulation model is developed based on a descriptive model. Before the simulation model can be developed all relevant information from service resources and processes are needed. For this reason the simulation model consists of two essential sub models.
On the one hand the description of resources is essential. Therefore information like the type of resource (e.g. human being, tools, vehicles or spare parts etc.) is necessary. Further the availability of the different resource types must be inserted into the model. For example the description of availability for human resources can be described with the help of parameters like working hours each day, allowed overtime hours each day, working hours at the weekend and expected days of illness and holidays. The availability of external resources which cannot be scheduled autonomously {15} is stochastically simulated by giving them an expectancy value. Beside the availability all resources must have a description of qualification and experiences (for human beings) or specification (for technical resources). Further the cost for each resource must be part of the input data.
On the other hand the second essential sub model is the description of service offers and processes which are executed by the service provider. The most important information which is needed for this description is the expected frequency of demands to simulate the process stochastically if processes occur unexpected (like breakdowns). But also demands which can be scheduled (like inspections or preventive maintenance) or which are regularly (like periodic maintenance activities) can be inserted into the model. Further relevant attributes of processes are required and must be part of the input data. Worth mentioning are attributes of service activities like “allowed reaction time” which define the latest start time or a time interval in which an execution of the service is possible without paying penalties. Besides this further time-related information is needed. The period of use for each resource during a service process must also be defined as well as the relatively start time for each resource in the process. To simulate realistic durations of a process an expectation value for possible delays should also be defined. In addition to costs, penalties (in €/hour)( in case of delayed execution of process) and further information like “possibility for subcontracting” as well as the information about required resources are needed. Hence each service offer needs the description of requested service resources. Thereby not only the quantitative information is needed but also the description of qualifications and specifications is essential. Further the interrelations between specifications and qualifications must be considered. E.g. if a service process needs an educated electrician with additional qualifications in the area of fall protection then these qualifications must be integrated into one skill profile. A combination of two service technicians (one with electrical qualification and another with experiences in industrial climbing) is not valid.
All described information (input data) is essential to simulate a realistic degree of capacity utilisation. With the help of intelligent algorithm for prioritisation of service orders and for allocation of resources a realistic virtual image of the daily operative business of a service provider can be created. On the basis of these simulation results the following questions can be answered and several sensitivity analyses can be calculated.
How many service resources are needed to meet the customer demands?
Which types of resources cause bottlenecks?
Which costs and penalties do providers have to expect?
Which quantified influences occur through the offered service level? (e.g. 95% vs. 97% availability)
Which effects result from a high or low qualification- and cooperation level?
How long is the average waiting time for customers or idle time for resources?
Which effect has the local service infrastructure through shorter or longer traveling times?


Conclusion

By means of the simulation model the CEO´s or head of services of providers are highly supported in their strategic decision making process. With the help of several sensitivity analyses deciders are able to find out their range of cost-efficient service capacities. Further the line-up of service infrastructures as well as an adequate qualification and cooperation level can be quantified. Additionally the financial risk can be evaluated and quantified. The model simulates a realistic resource usage which can be easily validated by experts through scopes and clearly arranged output data. By running several simulations the sensitivities analysis can be generated and interesting curves can be plotted. The following picture illustrates qualitatively an extract of the expected results to clarify which sensitivity analysis and outcomes are able to generate. . With the help of new and innovative algorithm providers are enabled to simulate their general new conditions (including the shift of risk and the resulting changes in resource allocation and prioritization). Target of this new innovative approach is not a work plan including a most homogenous workload for the service resources (which could be an economic criterion for the functional business models). But instead of that a realistic and economic allocation algorithm of resource to reach an efficient way to serve most of the service orders in time is implemented. Through this innovative approach the economic balance between cost caused by capacities and penalties can be quantified ex-ante for the first time by companies which are shifting their business models from a functional based to an availability based business model.WinServ is funded by the operational program “Regionale Wettbewerbsfähigkeit und Beschäftigung” of NRW from 2007-2013. The project is co-financed by EFRE.


Learning objectives
Simulation seems to be a promising approach to handle the complexity {16} and to support decision makers in the wind energy sector. The software “Matlab” in combination with “Simulink” is an adequate tool to model the described trade-off situation. However, you need experiences from the operative service activities and reliable input data for the simulation to receive meaningful results. If the input data are reliable the simulation model is a powerful tool for the strategic management.


References
{1} GWEC; Greenpeace International; DLR; Ecofys and the University of Utrecht (2012): Global Wind Energy Outlook 2012. Online available: http://www.gwec.net/wp-content/uploads/2012/11/GWEO_2012_lowRes.pdf
{2} Sawin, J. L.; Chawla, K.; Hinrichs-Rahlwes, R.; Macias G.E.; Angus, M.; Musolino, E. et al. (2013): Renewables 2013. Global Status Report. Hg. v. REN 21 Renewable Energy Policy Network for the 21st Century. Paris.
{3} Thomassen, P.; Ansorge, B.; Wienholdt, H. (2011): Creating synergies in the aftermarket: using the service network analysis for designing wind energy service networks. In: EWEA (Hg.): Scientific Proceedings European Wind Energy Conference. Brussels: EWEA, S.297-299.
{4} Enercon (2010): Windenergieanlagen. Partnerkonzept (EPK). Hg. v. Enercon GmbH. Aurich. Online available: http://www.enercon.de/p/downloads/Enercon_EPK_2010_deu.pdf
{5} General Electric (2011): GE Wind Energy GmbH: GE bietet produktionsbasierte Verfügbarkeitsgewährleistung auf Windenergieanlagen. Hg. v. General Electric. Salzbergen. Online available: http://w3.windmesse.de/windenergie/news/9385-ge-wind-energy-gmbh-ge-bietet-produktionsbasierte-verfugbarkeitsgewahrleistung-auf-windenergieanlagen
{6} Schuh, G.; Ansorge B. (2013): Key Dimensions oft he Transformation from Products to Services – A Framework following the Service-Dominant Logic; In: XXIII International RESER Conference Proceedings, o.S., Aix-en-Provence
{7} Korte, S.; Rijkers-Defrasne, S.; Zweck, A.; Knackstedt, R.; Lis, L. (2010): Hybride Wertschöpfung. Ergebnisse des Förderschwerpunktes. Hg. v. Zukünftige Technologien Consulting. VDI Technologiezentrum GmbH. Düsseldorf (Band Nr. 89). Online available: http://www.zukuenftigetechnologien.de/zt-consulting/publikationen.php?we_objectID=9377&st=508
{8} Lanza, G.; Behmann, B.; Peters, St. (2010): Verfügbarkeitswirksame technische Dienstleistungen. Potenzialanalyse bei KMU im Maschinen- und Anlagenbau. In: ZWF - Zeitschrift für wirtschaftlichen Fabrikbetrieb 105, 2010 (5), S. 421–425.
{9}Völker, Oliver (2012): Erbringungsorganisation hybrider Leistungsbündel, -Zugl.: Diss., Univ., Bochum.
{10} Gemmel, Paul; van Looy, Bart; van Dierdonck, Roland (2013): Service management. An integrated approach. 3. Aufl., Harlow: Pearson.
{11} Haller, Sabine (2012): Dienstleistungsmanagement. Wiesbaden: Springer Gabler.
{12} Möller, K.; Cassack, I (2008).: Prozessorientierte Planung und Kalkulation (kern-)produktbegleitender Dienstleistungen. In: Zeitschrift für Planung & Unternehmenssteuerung, Bd. 19, S. 159–184. Online available: http://cms.springerprofessional.de/journals/JOU=00187/VOL=2008.19/ISU=2/ART=51/BodyRef/PDF/187_2008_Article_51.pdf, zuletzt geprüft am 02.10.2013.
{13} Hodapp; W. (2009): Die Bedeutung einer zustandsorientierten Instandhaltung. Einsatz und Nutzen in der Investitionsgüterindustrie. In: Jens Reichel, Johannes Mandelartz und Gerhard Müller (Hg.): Betriebliche Instandhaltung. Heidelberg [u.a.]: Springer Berlin Heidelberg (VDI-Buch), S. 135–149.
{14} Chase, R. B.; Jacobs, F. R.; Aquilano, N. J. (2007): Operations management for competitive advantage with global cases. 11. Aufl. Boston: McGraw-Hill-Irwin (Operations and decision sciences)
{15} Corsten, H.; Gössinger, R. (2007): Dienstleistungsmanagement. 5. Aufl. Hg. v. Ralf Gössinger. München [u.a.]: Oldenbourg.
{16} Gudergan, G.; Ansorge, B.; Frombach, R. (2011): Exploring Interdependencies of Innovation Systems. In: Rollins College Henrique Correa (Hg.): Proceedings of the 22nd Annual POM Conference. Reno, Nevada, 2. May. Reno, Nevada: POMS, S. 131–142. Online available http://www.pomlearning.org/Reno/FullPapers/Fullpaper.htm