Naslov (srp)

Modeli za podršku odlučivanju na području poslovne logistike : doktorska disertacija

Autor

Pajić, Vukašin, 1994-

Doprinosi

Andrejić, Milan, 1983-
Kilibarda, Milorad, 1962-
Kaplanović, Snežana, 1972-
Nikoličić, Svetlana, 1962-

Opis (eng)

Business logistics stands as one of the intricate components within the supply chain due to its extensive involvement of participants and activities pivotal for a company's competitive edge. These activities encompass ordering, procurement, inventory management, distribution, etc. Notably, the efficacy of procurement and distribution processes can wield a substantial influence on a company's overall efficiency and competitiveness. Consequently, this doctoral dissertation centers its focus on these two logistics facets: procurement and distribution. The dissertation's objective is to introduce decision support models within the realm of business logistics. To fulfill this aim, two decision support models for procurement logistics and two for distribution logistics have been developed and proposed in this dissertation.When addressing procurement logistics, one of the pivotal activities is the selection of suppliers and order allocation. Consequently, the first model developed within this dissertation specifically tackles this challenge. Existing literature includes models related to supplier selection and order allocation. However, these models are primarily grounded in linear programming. Once the model is solved and the objective function is obtained, it essentially dictates from which supplier a specific quantity of goods should be ordered. The model developed in this dissertation takes a distinctive approach compared to those in the literature. In this model, the DEA method was initially applied to pinpoint only the efficient suppliers from the initial pool, subsequently subjected to further analysis and evaluation. The rationale behind employing the DEA method is to sidestep evaluating suppliers who lack efficiency and, consequently, to avoid ordering from such suppliers, which could detrimentally impact the efficiency of the ordering company. Following the implementation of the DEA method, in the second phase, the FUCOM method was applied to determine the criteria weights used for evaluating suppliers. In the third phase, the CoCoSo method was employed to rank the suppliers. Based on the results from this phase, in the last stage, a model for order allocation was developed and implemented. It can be concluded that the primary distinction of this model compared to those present in the literature lies in the fact that this model integrates MCDM methods with linear programming to address the issue of supplier selection and order allocation. Considering that procurement predominantly occurs at the international level, a crucial aspect of international trade involves negotiating international delivery terms based on Incoterms. These rules clearly delineate the responsibilities and obligations of the seller on one side and the buyer on the other. Additionally, these rules define the moment of risk and cost transfer from the seller to the buyer. Given the significance of choosing the appropriate delivery term, a model addressing this issue was developed in this dissertation. The model's objective was to conduct a survey to ascertain the factors influencing the choice of delivery terms, the most commonly used modes of transportation, critical factors in choosing delivery terms, and the most frequently utilized delivery terms. After conducting the survey and processing the results, criteria and variant solutions were identified and then evaluated to determine the optimal delivery term for the observed case. To ascertain the weights of the criteria used for evaluating variant solutions, the MEREC was applied. After determining the weights of the criteria, variant solutions were ranked using the MABAC method.Doktorska disertacija Vukašin PajićVAs previously emphasized, alongside procurement logistics, distribution logistics plays a crucial role in a company's operational efficiency and competitiveness. For this reason, two models have been developed to address issues related to distribution. In the execution of the distribution process, many companies opt to outsource this activity due to a lack of internal capabilities. However, choosing a suitable company for this process is not straightforward, particularly considering the complexities of today's business environment, marked by uncertainty, disruptions, and various global challenges. Companies entrusted with this process are commonly referred to as logistics providers, and depending on the scope of services offered, they are categorized as 3PL and 4PL providers. Based on the literature review, it was determined that there are numerous models related to the selection of 3PL providers. The primary objective of the model developed in this dissertation is to establish a foundation for decision support in the selection of resilient 4PL providers for delivering goods in e-commerce. In contrast to existing models in the literature, the developed model concurrently considers resilience, e-commerce dynamics, and the selection of 4PL providers. The described model is based on the application of fuzzy FUCOM, ET, RBT, and WASPAS methods. The fuzzy FUCOM method was applied to determine the criteria weights used in evaluating variants. Following the determination of weights, the ET and RBT methods were applied in the second phase of the model to establish the initial decision matrix. This matrix was then utilized in the third phase of the model, employing the WASPAS method. The advantage of employing these methods lies in their ability to handle incomplete assessments, providing flexibility not found in standard MCDM methods. The WASPAS method was then employed to rank variant solutions (4PL providers). When considering the distribution process, various performance indicators can be defined to gauge the efficiency of this operation. One such widely utilized indicator is OTIF. When negotiating distribution, particularly when outsourcing this process, the level of OTIF fulfillment holds paramount importance, as it sets the threshold for penalty imposition (if the distributor fails to execute distribution efficiently) or bonuses (if the distribution proves to be efficient). Bearing the aforementioned considerations in mind, the objective of the fourth model developed in this dissertation is to measure OTIF fulfillment and efficiency in the product distribution process. The developed model is based on the DEA method and has been tested in the context of spare parts distribution.

Opis (srp)

Poslovna logistika predstavlja jedan od najkompleksnijih segmenata lanca snabdevanja s obzirom da se sastoji od velikog broja učesnika i aktivnosti koje su jako važne za konkurentnost jedne kompanije. Te aktivnosti se prvenstveno odnose na poručivanje, nabavku, upravljanje zalihama, distribuciju, i dr. Od svih navedenih aktivnosti, efikasnost procesa nabavke i distribucije može najviše uticati na efikasnost poslovanja i konkurentnost jedne kompanije. Upravo iz tog razloga, predmet ove doktorske disertacije su ova dva segmenta logistike, nabavka i distribucija. Cilj disertacije jeste da predloži modele za podršku odlučivanju na području poslovne logistike. Kako bi se ostvario postavljeni cilj, u disertaciji su definisana dva modela za podršku odlučivanju na području logistike nabavke i dva modela za podršku odlučivanju na području logistike distribucije.Kada se govori o logistici nabavke, jedna od najznačajnijih aktivnosti jeste izbor dobavljača i alokacija poručivanja. Iz tog razloga, prvi model koji je razvijen u okviru disertacije, bavi se upravo ovom problematikom. U literaturi su prisutni modeli koji se odnose na izbor dobavljača i alokaciju poručivanja. Međutim, ti modeli su uglavnom bazirani na linearnom programiranju, gde jednom kada se reši model i dobije funkcija cilja, ona zapravo određuje od kog dobavljača je potrebno poručiti određenu količinu robe. Model koji je razvijen u ovoj disertaciji ima drugačiji pristup u odnosu na modele iz literature. Naime, u ovom modelu, prvo je primenjena DEA (engl. Data Envelopment Analysis) metoda kako bi se od početnog broja dobavljača izdvojili samo efikasni koji su potom dalje analizirani i vrednovani. Razlog za primenu DEA metode jeste to da se izbegne vrednovanje dobavljača koji nisu efikasni a samim tim i poručivanje od takvih dobavljača koje bi negativno uticalo na efikasnost rada same kompanije koja poručuje robu. Nakon implementacije DEA metode, u drugoj fazi, primenjena je FUCOM (engl. Full Consistency Method) metoda za utvrđivanje težina kriterijuma koji su korišćeni za vrednovanje dobavljača. U trećoj fazi, primenjena je CoCoSo (engl. Combined Compromise Solution) metoda za rangiranje dobavljača. Na osnovu rezultata iz ove faze, u poslednjoj fazi, razvijen je i primenjen model za alokaciju poručivanja. Može se zaključiti da je glavna razlika ovog modela u odnosu na one koji su prisutni u literaturi činjenica da ovaj model kombinuje MCDM (engl. Multi Criteria Decision Making) metode sa modelom linearnog programiranja kako bi rešio problem izbora dobavljača i alokacije poručivanja. S obzirom da se nabavka uglavnom odvija na međunarodnom nivou, jedan od sastavnih delova međunarodne trgovine jeste ugovaranje međunarodnih uslova isporuke iz Incoterms-a (engl. International Commercial Terms). Ova pravila jasno definišu odgovornosti i obaveze prodavca sa jedne strane i kupca sa druge. Takođe, ova pravila definišu i trenutak prelaska rizika kao i troškova sa prodavca na kupca. Imajući u vidu značaj izbora adekvatnog pravila isporuke, model koji se bavi upravo ovom problematikom razvijen je u ovoj disertaciji. Cilj modela bio je da se prvo sprovede anketno istraživanje kako bi se odredili faktori koji utiču na izbor pravila isporuke, najčešće korišćeni vidovi transporta, faktori koji su od presudnog značaja prilikom izbora pravila isporuke, kao i to koja se pravila isporuke najčešće koriste. Nakon sprovedenog anketnog istraživanja i obrade rezultata, izdvojeni su kriterijumi i varijantna rešenja koja su potom vrednovana sa ciljem da se za posmatrani primer izabere optimalno pravilo isporuke. Kako bi se odredile težine kriterijuma koji su korišćeni za vrednovanje varijantnih rešenja primenjena je MEREC (engl. Method based on the Removal Effects of Criteria) metoda. Nakon određivanja težina kriterijuma, varijantna rešenja su potom rangirana primenom MABAC (engl. Multi-Attributive Border Approximation Area Comparison) metode.Doktorska disertacija Vukašin Pajić Kao što je već napomenuto, pored logistike nabavke, logistika distribucije predstavlja jedan od ključnih faktora efikasnosti poslovanja i konkurentnosti jedne kompanije. Iz tog razloga, definisana su dva modela koja rešavaju probleme povezane sa distribucijom. Naime, prilikom realizacije procesa distribucije brojne kompanije se odlučuju da autsorsuju ovu aktivnost usled nedostatka sopstvenih kapaciteta. Međutim, izbor kompanije kojoj će se poveriti ovaj proces nije jednostavan, naročito ukoliko se uzmu u obzir karakteristike današnjeg poslovnog okruženja (neizvesnost, poremećaji, brojni izazovi prisutni na svetskom nivou, i dr.). Kompanije kojima se ovaj proces prepušta nazivaju se logistički provajderi i u zavisnosti od stepena obuhvatnosti pružene usluge najčešće se govori o 3PL (engl. Third-party Logistics) i 4PL (engl. Fourth-party Logistics) provajderima. Kako je pregledom literature utvrđeno da postoji veliki broj modela koji se odnose na izbor 3PL provajdera, cilj modela razvijenog u ovoj disertaciji bio je da se stvori osnov za podršku odlučivanju prilikom izbora rezilijentnih 4PL provajdera za isporuku robe u e-trgovini. Za razliku od prisutnih modela u literaturi, razvijeni model istovremeno uzima u obzir rezilijentnost, e-trgovinu i izbor 4PL provajdera. Opisani model baziran je na primeni fuzzy FUCOM, ET (engl. Evidence Theory), RBT (engl. Rule-Based Transformation) i WASPAS (engl. Weighted Aggregates Sum Product Assessment) metoda. Fuzzy FUCOM metoda je primenjena kako bi se odredile težine kriterijuma koji su korišćeni prilikom vrednovanja varijanti. Nakon određivanja težina, u drugoj fazi modela primenjene su ET i RBT metode kako bi se odredila početna matrica odlučivanja koja je potom korišćena u trećoj fazi modela, u WASPAS metodi. Prednost primene ovih metoda ogleda se kroz mogućnost nepotpune procene, što nije slučaj kod standardnih MCDM metoda. Primenom WASPAS metode izvršeno je rangiranje varijantnih rešenja (4PL provajdera). Prilikom realizacije procesa distribucije moguće je definisati mnogobrojne indikatore performansi koji će dati sliku o efikasnosti ovog procesa. Jedan od tih indikatora, koji se često primenjuje u praksi, jeste OTIF (engl. On-Time In-Full). Prilikom ugovaranja distribucije, naročito kada se ovaj proces autsorsuje, jako je bitan nivo ispunjenosti OTIF-a s obzirom da on definiše prag naplate penala (ukoliko distributer ne izvršava distribuciju efikasno) ili bonusa (ukoliko je distribucija efikasna). Imajući prethodno navedeno na umu, cilj četvrtog modela koji je razvijen u ovoj disertaciji jeste merenje ispunjenosti OTIF-a i efikasnosti u procesu distribucije proizvoda. Razvijeni model je baziran na DEA metodi, i testiran na primeru distribucije rezervnih delova.

Opis (srp)

Saobraćajno inženjerstvo - Poslovna logistika i špedicija / Traffic engineering - Business logistics and freight forwarding Datum odbrane: 26.04.2024.

Jezik

srpski

Datum

2024

Licenca

Ovo delo je licencirano pod uslovima licence
Creative Commons CC BY-NC-ND 3.0 AT - Creative Commons Autorstvo - Nekomercijalno - Bez prerada 3.0 Austria License.

http://creativecommons.org/licenses/by-nc-nd/3.0/at/legalcode

Predmet

OSNO - Opšta sistematizacija naučnih oblasti, Logistika

Poslovna logistika, nabavka, distribucija, trgovina, Incoterms, višekriterijumsko odlučivanje

OSNO - Opšta sistematizacija naučnih oblasti, Logistika

Business logistics, procurement, distribution, trade, Incoterms, multi criteria decision-making (MCDM)