Napovedovanje pretovora blaga za podporo planiranja aktivnosti v Luki Koper, d.d.

Zarnec, Mirjana

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Napovedovanje pretovora blaga za podporo planiranja aktivnosti v Luki Koper, d.d.
Authors:	ID Zarnec, Mirjana (Author) ID Dragan, Dejan (Mentor) More about this mentor...
Files:	MAG_Zarnec_Mirjana_2014.pdf (5,28 MB) MD5: 0A1896C92E9E6F54618A095E2645A158
Language:	Slovenian
Work type:	Master's thesis/paper
Organization:	FL - Faculty of Logistic
Abstract:	Za uspešnost poslovanja podjetij je pomembno, da se poslovne odločitve na operativnem, taktičnem in strateškem nivoju sprejemajo na podlagi napovedi poslovanja. Napovedi so lahko kvantitativne ali kvalitativne. Za katero vrsto napovedi se bomo odločili, je odvisno od vrste in obsega podatkov, ki jih imamo na voljo. Prav tako pa na odločitev vpliva dolžina napovedi, ki jo potrebujemo. V gospodarstvu, se na operativnem nivoju najpogosteje uporabljajo kratkoročne napovedi, na taktičnem nivoju srednjeročne napovedi ter dolgoročne napovedi na strateškem nivoju. Za sprejemanje odločitev na strateškem nivoju potrebujemo dolgoročne napovedi, za kar so primernejše kvalitativne metode. Za sprejemanje kratkoročnih odločitev so primernejše kvantitativne metode, s katerimi načeloma lahko predvidimo zelo natančno povpraševanje v naslednjih nekaj obdobjih. V magistrskem delu smo analizirali podatke o pretovoru na primeru Luke Koper, d. d. Podatki, ki smo jih uporabili za analizo, so javno dostopni na spletu. Za analizo smo izbrali mesečne podatke o skupnem pretovoru, pretovoru kontejnerjev in pretovoru RO-RO (roll-on roll-off). Za tehnike napovedovanja smo izbrali multiplo regresijsko analizo, eksponentno glajenje in SARIMA model (model sezonskih avtoregresijskih integriranih drsečih sredin). Za vsako od treh izbranih skupin pretovora smo izbrali najboljši model glede na RMSE (koren povprečne kvadratne napake) in MAPE (odstotek povprečne absolutne napake) ter na koncu primerjali rezultate vseh treh tehnik, ki smo jih uporabili za napovedovanje. Med izbranimi modeli ne prihaja do večjih razlik pri napovedih. Ugotavljamo, da lahko z vsemi tremi modeli izdelamo približno enako dobre kratkoročne napovedi. V letu 2014 pričakujemo povečanje pretovora za vse tri blagovne skupine, ki smo jih vključili v analizo. Največjo rast, 13,5 %, pričakujemo pri pretovoru kontejnerjev. Za skupen pretovor analize kažejo na 4,5 % povečanje pretovora, kar je 0,5 % manj, kot so napovedali strokovnjaki Luke Koper.
Keywords:	napovedovanje pretovora, multipla regresijska analiza, eksponentno glajenje, SARIMA, pomorski transport
Place of publishing:	Celje
Publisher:	[M. Zarnec]
Year of publishing:	2014
PID:	20.500.12556/DKUM-44177
UDC:	656.6
COBISS.SI-ID:	512574525
NUK URN:	URN:SI:UM:DK:FQACB56D
Publication date in DKUM:	22.07.2014
Views:	2336
Downloads:	290
Metadata:
Categories:	FL
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Secondary language

Language:	English
Title:	Throughput forecasting to support the planning activities in the port of Koper
Abstract:	For the business success of the companies it is important that their decisions are based on the experts forecasts at operative, tactical and strategic level. Forecasts can be qualitative or quantitative. Which type of forecast we use, depends on the type and the range of the data available. Also, the decision is affected by the length of the needed forecast. In the economy, most commonly used forecasts at the operational level are short-term forecasts, at the tactical level medium-term forecasts and long-term forecasts at the strategic level. For long term forecasts are more relevant quantitative analysis and the decision making at the strategic level requires long term forecasts. For short term decision making, quantitative forecasts provide better results. In principle predicting demand with quantitative analysis can be very accurate for next few periods. In the master thesis we have analyzed data on the throughput for Port of Koper. Data used for the analysis is available online. For the analysis we have collected monthly data of total throughput, container throughput and RO-RO (roll-on roll-off) throughput. Multiple regression analysis, exponential smoothing and SARIMA (seasonal autoregressive integrated moving average) model were used for throughput forecasting. For each commodity group we estimated a model and chose the best one according to RMSE (root mean squared error) and MAPE (mean absolute percentage error). In the forecasting chapter we have compared results of all three estimated models. Results obtained with three different models are approximately the same. All three models can be used to provide strong short-term forecasts. We are expecting increase of throughput for all analyzed commodity groups. Comparing to the last years’ throughput, the highest increase is predicted for the container throughput, which is estimated to be 13,5 % higher. According to our analysis 4,5 % increase is expected for the total throughput, which is 0,5 % less as the Port of Koper’s estimation.
Keywords:	throughput forecasting, multiple regression analysis, exponential smoothing, SARIMA, shipping

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back