Veri Zarflama Analiziyle Hizmet Etkinliği Ölçümü: Diyarbakır DİSKİ Örneği

Yıl 2014, Sayı: 31.1, 179 - 187, 01.01.2014

Mehmet Karahan Recep Akdağ

Öz

Yerleşik yaşamın en temel gereksinimlerinden birisi altyapıhizmetlerinden olan su ve kanalizasyon hizmetleridir. Özellikle günümüz dünyasında nüfus artışı, su kaynaklarının kirletilmesi, bilinçsiz kullanım, ülke yönetimlerince suyun politik bir araç olarak kullanılmasıgibi, birçok nedenden dolayısu yönetimiyle ilgili krizler sıkça gündeme gelmektedir. Aynı şekilde kullanılan suyun atık haline geldikten sonra toplanması, iletilmesi ve arıtılarak çevreye zarar vermeden alıcıortama sunulmasıda çevre bilincine katkıanlamında oldukça önemlidir. Bu bağlamda, günümüzde su ve kanalizasyon hizmetlerinin sunulmasıesnasında ekinlik, verimlilik, kalite, insan ve çevre odaklılık gibi konular üzerinde çalışmalar yapılan önemli konulardır. Bahis konusu bu hizmetlerin kamu, özel veya kamu-özel ortaklığıtarafından yerine getirmesi gerekliliği, günümüzde tartışılan konuların önünde gelmektedir. Dünyanın birçok ülkesinde, Türkiye’de de olduğu gibi su ve kanalizasyon hizmetleri kamu eliyle, yani yerel yönetimler tarafından yerine getirilmektedir. Türkiye’de su ve kanalizasyon hizmetleri, kent merkezlerinde il belediyelerince, büyükşehirlerde bağlıkuruluşlar, kent merkezi dışında ise il özel idareleri tarafından karşılanmaktadır. Çalışmada; Diyarbakır Büyükşehir Belediyesi’ne bağlıkuruluşolan Diyarbakır Su ve Kanalizasyon İdaresi’nin yıllara göre hizmet etkinliği Veri Zarflama Analizi ile ölçülmüş, etkinliğin yıllara göre değişimi karşılaştırmalıolarak incelenmiştir. Etkin olmayan yıllar tespit edilerek sorunların nedenleri analiz edilmiştir. Etkinlik ölçümü su ve kanalizasyon olmak üzere iki hizmet alanına göre, 2000-2012 yıllarıiçin yapılmışve böylelikle her yılın etkinlik düzeyi birbiriyle karşılaştırılmıştır. Veri zarflama analizi çıktıya yönelik BCC metoduyla yapılmışve girdi değişkenleri olarak; toplam personel sayısı, içme suyu şebeke uzunluğu, atık su şebeke uzunluğu, kente verilen su miktarıile ilgili veriler, çıktıdeğişkenleri olarak da toplam abone sayısı, faturalandırılan su ve atık su miktarıile ilgili verileri kullanılmıştır. Çalışma sonuçlarına göre, 2003, 2005 ve 2008 yıllarında kurumun etkinlik düzeyinin düşük olduğu belirlenmiştir. 2003 yılına ait çıktılarda, girdilerden daha fazla artışolduğu gözlenmiş, 2005, 2007 ve 2008 yıllarında çıktıların, girdilerden daha az artışgösterdiği tespit edilmiştir. Çalışmada bir işletmenin geçmişyıllarına ait girdi ve çıktıverileri kullanarak VZA metodu ile etkinlik düzeyi belirlenmişve bu sayede işletmenin ölçek, kaynak kullanımıve üretim etkinliklerinin değerlendirilmesi konularına önemli katkılar sağlamasıhedeflenmiştir. Ayrıca işletmenin gelecek yıllarda yapacağıplanlamalarında da elde edilen bu bilgilerin yol gösterici olmasıbeklenmektedir.

Anahtar Kelimeler

Etkinlik analizi, Hizmet etkinliği, Veri zarflama analizi

Measuring of Service Efficiency by Data Envelopment Analysis: Sample of Diyarbakir DISKI

Öz

The purpose of this study is to measure the service efficiency of DISKI, which is affiliated with Diyarbakir metropolitan municipality and serve as water and sewerage administration, by years using Data Envelope Analysis DEA and then to determine the change in the efficiency by years and finally to develop some proposals for the solution of identified problems for inefficient years. In the literature review, it has been realized that there are many studies comparing efficiency level of municipalities in terms of water and sewerage services together with other services. In these studies various municipalities had been compared by DEA method and comparative efficiency analyses had been performed. In some oversea studies, again efficiency research had been performed by using DEA De Borger and Kerstens, 1996; Poister and Streib, 1999; Prieto and Zofio, 2001; Sousa and Stosic 2005; Loikkanen and Susiluoto 2006; Balaguer-Coll etc., 2007; Afonso and Fernandes, 2008 . Some domestic studies, by Çağlar 2003 , Doğan 2006 , Kaplan etc. 2006 , Güneş and Akdoğan 2007 , İlkay and Doğan 2009 are also remarkable. In the literature, we can give especially some examples, Tupper and Resende 2004 -Brazil, Woodbury and Dollery 2004 -Australia Bağdadioğlu and Cumhur 2010 -Turkey, as studies that compare municipalities by measuring efficiency of services by water and sewerage administrations of municipalities. In this studies efficiency researches had been done by DEA and solutions developed for the correction of inefficient processes. In application phase, efficiency analysis of water and wastewater services by DISKI was carried out to include 13 years, between 2000 and 2012, by DEA. During literature review, we noticed that similar variables were used in previous studies to determine input and output variables used for efficiency analyses Bağdadioğlu and Cumhur, 2010; Cağlar, 2003; Doğan, 2009; Kaygısız, 2011 . These variables are shown in the following Table 1. Table 1. Model Variables Input Variables Output Variables Total staff number The length of drinking water network km. The length of wastewater network km. The length of water supplied to the city m3 Total number of subscribers The amount of billed for water and wastewater m3 Input and output data variables used for DEA are collected from annual activity reports of DISKI. In this study, DEA was carried out by output-oriented BCC to see the maximum number of output that can be obtained as a result of the combination of four inputs and two outputs. We obtained solution values of BCC model by using Win4DEAP Version1.1.3. software. In Table 2, according to BCC model “pure efficiency value” of efficient years is marked as 1, while “pure efficiency value” of inefficient years are marked as 4, 6, 9, which are namely 2003, 2005, and 2008. Scale efficiency increased in 2003 which means that the increase in the amount of output is greater than the increase in the amount of input. In 2005, 2007 and 2008, scale efficiency decreased because the amount of output is less than the amount of input. Among efficient years, the reference intensity of years: 2006 three times, 2002 and 2009 two times, 2000, 2001 and 2004 years one time. The values in reference set column mean to take reference from which year and how much percentage for non-efficient years to be efficient or to be able to come to efficiency border. For example, fourth DMU 2003 year can be efficient or come to efficiency border taking reference at percentages of 0.287, 0,622 and 0.091 to 7th, 3th and 2nd DMUs respectively. When current values of in non-efficient years and percentages in years taken reference multiply we can calculate to be efficient required values in these years. In our example, required value of total number of subscribers for fourth DMU is obtained as following. 143,565 x 0.287 + 120,893 x 0.622 + 112.436 x 0.091 = 126,631 Fourth DMU’s 2003 year value of current total number of subscribers is 125, 376. To consider total number of subscribers in this year as efficient, there must be an increase of 126,631 - 125,376 = 1.255 units in the number of subscribers. In this study, efficiency of input and output values regarding water and wastewater services by DİSKİ in the years 2000 to 2012, which is a local administration unit, was analyzed. As a result of the analysis, DISKI has been found to be efficient in 10 of 13 years and inefficient in three years, namely 2003, 2005 and 2008. When evaluating non-efficient years on input basis, in 2008 the number of staff, in 2003 the amount of drinking water network, in 2003 and 2005 the amount of water supplied to the city are higher, so it has been concluded that the inputs are used excessively. When evaluating on output basis by years, level of output is low which means that total number of subscribers and the amount of water and wastewater billed in all 2003, 2005 and 2008 years are low, so it can be concluded that desired output level could not be attained. In the result obtained at inputs, especially in 2008 with an excess of staff and in 2003 with an excess of the amount of water supplied to city is remarkable. When the amount of water supplied is over but the amount of water and wastewater billed is inadequate, the situation indicates the presence of illegal consumption and water network losses. On the other hand in 2003 the amount of excess water supplied to city show a decrease in 2005 and reached zero level in 2008 the amount of water and wastewater billed in 2005 and 2008 each year approximately 50% has decreased. Despite this, the total number of subscribers in 2005 and 2008 seems to be not enough. In this case, while supplied water is enough and even excess and there are improvements in the amount of billed water and wastewater but total number of subscribers is at inadequate level, so this indicates there can be an increase in the number of consumers who are not subscriber. Input and output values in these years have reached an efficient level in ensuing years. However when determining the efficiency border, efficiency levels of 2010, 2011 and 2012 years also need to be analyzed to reveal, taking as reference mostly 2006 and previous years and taking as reference only 2009 after 2008. When scale efficiency is taken into consideration, increase in the amounts of output in 2003 is more than increase in the amounts of input and the opposite situation increase in the amounts of output in 2005, 2007 and 2008 is less than the amounts of input that has been determined. This result in terms of efficient management of resources, compared to other years of 2003 has showed more efficiency. As shown in this study, we can obtain information about efficiency level of each year with DEA using values of input and output belonging to previous years. It is evident that the results of this information can be very useful for businesses during the evaluation of resource utilization and production efficiency of the future years. As a result, this study which was carried out on the data of input and output variables of previous years about only one business proved that conclusions obtained by comparing the efficiency levels of previous years of businesses will make significant contributions to business in making political decisions about scale efficiency, production, and marketing etc.

Anahtar Kelimeler

Efficiency Analysis, Data Envelopment Analysis, Service Efficiency

Kaynakça

• Afonso, A. ve Fernandes, S. (2008). “Assessing and explaining the relative efficiency of local government”. The Journal of Socio-Economics, 37(2008) 1946-1979.
• Akal, Z. (2000). İşlemlerde Performans Ölçüm ve Denetimi Çok Yönlü Performans Göstergeleri, MPM, No: 473, Ankara.
• Atan, M. ve Çatalbaş, G. K. (2005). “Bankacılıkta Etkinlik ve Sermaye Yapısının Bankaların Etkinliğine Etkisi”, 7. Ekonometri ve İstatistik Sempozyumu, 26-27 Mayıs 2005, İstanbul Üniversitesi, İstanbul.
• Bağdadioğlu, C. ve Cumhur Y. (2010). “Turkiye Şehir Suyu Arz Sektöründe X-Verimsizliğinin Ölçülmesi”, Hacettepe Üniversitesi İ.İ.B.F. Dergisi, Cilt 28, Sayı 1, 205-223.
• Balaguer-Coll, M.T., Prior, D., Tortosa-Ausina, E. (2007). “On the determinants of local government performance: A two-stage nonparametric approach”, European Economic Review, 51(2007) 425-451.
• Banker, R. D. (1984). “Estimating Most Productive Scale Size Using Data Envelopment Analysis”, European Journal of Operational Research, 17, 35-44.
• Banker, R. D., Cooper W. W., Seiford L. M., Thrall R. M., Zhu, J. (2004). “Returns to Scale in Different DEA Models”, European Journal of Operational Research, 154, 345–362.
• Boussofiane, A., Dyson, R., Rhodes, E. (1991). “Applied Data Envolopment Analysis”, European Journal of Operational Research, 2(6): 1-15.
• Bowlin, W. F. (1998). “Measuring Performance: An Introduction to Data Envelopment Analysis (DEA)”, The Journal of Cost Analysis, 3–27.
• Charnes, A., Cooper, W.W., Rhodes, E. (1978). “Measuring the Efficiency of Decision Making Units”, European Journal of Operation Research, 2, 429-444.
• Cooper, W.W., Seiford, L. M., Zhu, J. (2004). Handbook on Data Envelopment Analysis, Kluwer Academic Publishers, Boston.
• Çağlar, A. (2003). “Veri Zarflama Analizi ile Belediyelerim Etkinlik Ölçümü”, Yayınlanmamış Doktora Tezi, Hacettepe Üniversitesi Fen Bilimleri Enstitüsü İstatistik Ana Bilim Dalı, Ankara.
• DİSKİ (2000-2012). Faaliyet Raporları, Diyarbakır Büyükşehir Belediyesi Yayınları, Diyarbakır.
• Güran, M. C. ve Cingi, S. (2002). “Devletin Ekonomik Müdahalelerinin Etkinliği”, Akdeniz Üniversitesi İ.İ.B.F. Dergisi, 3, 56 -89.
• İlkay, M. S. ve Doğan, N. Ö. (2009). “Veri Zarflama Analizi İle Kapadokya Bölgesindeki Belediyelerin Etkinlik Ölçümü: 2004 ve 2008 Yıllarına İlişkin Bir Karşılaştırma”, Erciyes Üniversitesi İ.İ.B.F. Dergisi, 32, 191-218.
• Kaygısız, G. ve Giringer, N. (2011). “Maliyet Etkinlik Analizi: Türkiye’deki Büyükşehir Belediyelerinde Uygulama”, Eskişehir Osmangazi Üniversitesi İ.İ.B.F. Dergisi, 6(2), 309-342.
• Prieto, A. M. ve Zofio, J. L. (2001). “Evaluating Effectiveness in Public Provision of Infrastructure and Equipment: The Case of Spanish Municipalities”, Journal of Productivity Analysis, 15, 41-58.
• Ramanathan, R. (2006). “Data Envelopment Analysis for Weight Derivation and Aggregation in the Analytic Hierarchy Process”, Computers & Operations Research, 33, 1289–1307.
• Seiford, L. M. ve Thrall, R.M. (1990). “Recent developments in DEA”, Journal of Econometrics. 46, 7-38.
• Sousa, M. ve Stosic, B. (2005). “Technical Efficiency of the Brazilian Municipalities: Correcting Nonparametric Frontier Measurements for Outliers”, Journal of Productivity Analysis, 24, 157-181.
• Tavares, G. (2002). “A Bibliography of Data Envelopment Analysis (1978-2001)”, Rutcor Research Report, 01-02.
• Tupper, H. C. ve Resende, M. (2004). “Efficiency and regulatory issues in the Brezilian water and sewage sector: an empirical study”, Utilites Policy, 12(2004) 29-40.
• Woodbury, K. ve Dollery, B. (2004). “Efficiency measurement in Australian Local Government: The Case of New South Wales Municipal Water Services”, Review of Policy Research, 21(5): 615-636.
• Yolalan, R. (1993). İşletmelerarası Göreli Etkinlik Ölçümü, Ankara, MPM Yayınları, Yayın No: 483.
• Yun, Y. B., Nakayama H., Tanino T. (2004). “A Generalized Model for Data Envelopment Analysis”, European Journal of Operational Research, 157, 87–105.