PORT OPERATIONS SIMULATION MODEL

This work describes a simulation model of port operations for a port that serves oil platforms located on the Brazilian coast. Experiments were made to test different operational and strategic policies. In addition, issues related to demand and fleet profile and their effects on system performance were addressed.

The operation of supplying platforms requires large investments and has very high operating costs, especially with regard to per diems paid to support vessels and the operation of the port that serves these vessels. In this system, the efficient use of resources is essential.

It is in this scenario that the need arises to develop instruments capable of dealing with this complexity, and through which it is possible to determine more precisely the needs for capacity expansion due to the increase in demand, as can be seen in [2 ]. In addition, such instruments allow evaluations of possible alternatives in the arrangement of current resources and even in the evaluation of the effects of changes in the modus operandi and in the technology used in this system.

This work reports some aspects of the development of a tool designed to support the capacity planning of a port terminal to support off-shore operations. The technique employed was computational simulation, which is applied when the relations that make up the system under study are not simple enough for optimizing methods to be used. In addition, it allows studying the operation of systems such as the port in question, the way it currently works and also the way it could come to work, exploring the concept of what-if analysis. Such an approach becomes very useful when studying the adoption of new technologies, when experimenting with new resource arrangements and when the cost of implementing a new process (implementing and observing its performance) is too high to assume the risk of failure.

II - MODELING

The computational model, developed with the ARENA software, simulates the operation of a port that serves vessels that transport supplies to oil exploration and production platforms. Vessels are the “customers” of the port, which will have a longer or shorter service time, depending on the amount and type of resources available, as well as the availability of materials required by the vessels at the time of their arrival.

VESSELS

Vessels are divided into categories that differ according to some basic characteristics such as arrival rate, size, etc. and that determine greater or lesser use of port resources. The categories are:

• Supplies: vessels that use cranes, forklifts and carts to load and unload their materials. They also supply non-solid cargo such as water, diesel and mud, through hoses connected to the mooring piers.

• Bulk carriers: vessels that predominantly carry non-solid cargo, in bulk, using hoses and not cranes.

• Exclusive: vessels that have operations that are different from the others. Dedicated vessels are normally moored longer at the piers, however using cranes and hoses with much less intensity.

THE MATERIALS

Materials, as we have seen, can be classified into solids and non-solids. The latter use hoses to be loaded and the others use cranes, forklifts and trailers. Among the non-solid materials are water, diesel, barite mud, bentonite and cement.

Solid materials were basically divided into two classes, called “deck” and “tubes”. Deck material is that found in the form of containers, metal baskets, skids and other forms that are carried on the deck of the vessel. The tubes are also transported on deck, but differ from the material on the deck due to the time required to load and unload a trailer: as they are more difficult to handle, the operation takes longer.

THE RESOURCES

Trailers – There is a pool of trailers, that is, an area where trailers are concentrated, where they are parked when not in use. As they are requested, the trucks begin to operate, moving through all sectors of the port. If the number of trucks is insufficient to meet the number of requests, these (the requests) enter a queue and wait for the release of a truck.

Cranes – Cranes, or servers, handle cargo from vessels to trailers, or vice versa. The distribution that represents the operating time of these servers was calculated considering the complete loading (or unloading) of a trailer.

Berths – These are the places where the vessels dock. The number of berths has an influence on the average loading time of vessels because the more berths there are, the more vessels will be advancing the loading of non-solid materials. It should be noted that there may be more cradles than cranes.

LOCATIONS OF OPERATION

The materials unloaded/loaded on the vessels are destined/originated from certain locations in the port area. These locations may be further away or closer to the pier, where the vessels are located. If they are close, the travel time for a trailer from this location to the pier will be shorter and, therefore, the time to unload/load the entire vessel will also be shorter, which may require 10 to 30 trailers. On the contrary, if the place of operation is very far from the pier, the longer the trailers will be moving and the lower the general availability of trailers, which operate in a pool.

The operating locations (figure 1) are the deck pre-shipment, tube pre-shipment, contractors, tube park, and retroport. The furthest place is the tube park, about 120 minutes from the pier and the closest is the pre-boarding deck, which is 5 minutes away.

III - EXPERIMENTATION AND OBTAINED RESULTS

After modeling the system, the model was validated, which, according to [1] and [3], is essential for carrying out experiments that lead to representative responses for the system under study. After the validation, a series of experiments was carried out, of which 2 were selected as an example. They try to verify the behavior of the port system when subjected to changes in demand (arrival of ships) and in the fleet of vessels operating in the port, respectively.

Experiment A – The main objective of this experiment refers to the study of investment alternatives that aim to adapt the port's capacity to the possible increase in the number of platforms served. As an immediate consequence of this increase, the number of vessels arriving daily at the port tends to increase. For this, it is necessary to structure itself so that the level of service (assistance time) is not compromised. The alternatives analyzed were:

• Alternative 1: Remove exclusive vessels from the port, thus freeing up more resources for other vessels.

• Alternative 2: Double the number of servers (equipment for handling loading and unloading from vessels to trailers). Cranes are currently used, but through the model one can consider the installation of different equipment such as overhead cranes, which could load and unload several vessels simultaneously.

• Alternative 3: Similar to alternative 2, but also including the doubling of berths, that is, doubling the number of vessels that can simultaneously be supplied with water, diesel, bulk and fluids.

Figure 2 shows the average length of stay of supply vessels for the current situation and for each of the alternatives, using different arrival rates in the experiments. Alternative 1 (withdrawal of exclusive vessels from the port), in relation to the current situation, presents a significant reduction in the average length of stay of supply vessels only from a 15% increase over the current arrival rate. Increasing the arrival rate by 5% and 10%, the average time reduction does not exceed 1 hour.

Experiment B – This experiment aims to evaluate the effect of changing the fleet profile on port operations. The experiments carried out evaluated the operation with a homogeneous fleet of 380m2, that is, all supply vessels were converted to a standard area.

Basically, the same assumptions as in the previous experiment were adopted, with the exception of the arrival rate of supply vessels, which was corrected to adapt to a homogeneous fleet with a deck area of ​​380 m2. The arrival rate of this fleet increased to 4,57 vessels/day.

In figure 3 we can see that only after the 15% increase in the arrival rate of the vessels does the difference in the average stay time of the supply vessels become representative, with a reduction of 6,8h, in relation to the situation current. For a significant increase in demand (30%), a difference of 31 hours in the average length of stay was obtained.

IV - CONCLUSION

The experiments showed that the simulation of port operations allowed the identification of the port's saturation point, that is, until what demand the port could meet a certain level of service. In addition, it was possible to identify whether there was any advantage in moving exclusive vessels to another port.

Through the model, the possibility of restructuring the fleet profile was also tested. The result indicated that this change would not be effective for a demand increase of up to 10%. The other experiments sought to answer questions related to the investments needed to improve the level of service. The technological characteristics of the equipment were evaluated, as well as the changes in the operational characteristics of the system. The sizing of resources was seen in order to contemplate not only technological issues, but also the amount to be used.

In general, the port operations simulation model contributed to the study of numerous operational and even strategic situations. With this, the decision maker has information that enables an adequate sizing of the port.

V - BIBLIOGRAPHY

BANKS, J.; CARSON II, JS; NELSON, BL, Discret-Event System Simulation, 2nd ed., Prentice-Hall, 1996.

DARZENTAS, J.; SPYROU, T., Ferry Traffic in the Aegean Islands: A Simulation Study, Journal of the Operational Research Society, V.47, 1996, pp. 203-216.

LAW, AM, KELTON, WD, Simulation Modeling & Analysis. 2nd ed., McGraw-Hill, 1991.

SYSTEMS MODELING CORPORATION, ARENA User's Guide, 1996.

Authors: Leonardo Lacerda, Eduardo Saliby, Paulo Nazario and Marcelo Lara