SCM Optimization

Port Location Optimization in Oil and Gas Business

How Chose the best location of a Supply base

The Object of the problem is to study the usage and cost of the introduction of two new greenfield transshipment service ports in the Southern Caribbean/Guianas Oil and Gas service network. network shown below.

We are presently testing the Medium service to publish articles and content around Supply Chain issues in the Caribbean and specifically with the Oil and Gas business. This article is for all intent and purpose a Lorem Ipsum piece. I am a semi retired engineer and analyst who worked most of his career in the Oil and Gas service sector in Trinidad and Tobago.

Rigs and Supply Boats idle offshore Trinidad

This type of problem is sometimes referred to as the “Candidate Transshipment” problem that is illustrated as shown where we have material moving from a source to different customer groups via an intermediate set of transshipment points between the source and the destination.

• This write up does not use actual data because it is a ‘proof of concept’ for this port study.

https://www.instagram.com/p/COvP43krskv/

There are often many parameters that play a factor in the selection of these intermediate points. The Objective is to minimize overall transportation costs for a set of demand and supply constraints using an MILP[1] problem as the product traverse the source plant to the customers on the other side. Some other constraints typically are:

Different Capacities and costs of at the Transshipment points. Different costs at various nodes and over the arcs (transportation paths) of the network.

Using Excel Solver for Network Optimization

Level of Service Requirements (see…) and any conversion between LOS and cost. A minimum or a maximum required number of Intermediate Distribution Candidates from the full set.

Another metric that must be satisfied for the required solution that can be measured by the model during run time. For example — you may want to keep the minimum number of Ton-miles (or BlkU-miles) below a certain level or even optimize by this rather than cost.

The Network

TK We are going to explore what introducing two extra nodes in a 3 node shipping network, transshipping goods using a standardize unit referred to as BlkU or Bulk Unit, this can be anything that is typically shipped offshore — baskets, tubulars, pallets etc., but it is standardized into one unit as we are investigating a more generalized network.

Sourcing

The model will use source product coming from three possible sources:

• Zhanjiang, China
• Houston, US
• Macae, Brazil

Not much will analyzed from the sourcing options but it will give us a place to expand the model for further optimization where we could incorporate lead time factors which is important for longer shipping times where the variation from the normal can have a significant impact on source selection. We will have the option to open/close sources and vary supply capacities of the sources.

To get the Data from our table for the locations and their costs we use the following R Code:

ports <- data.frame(plocations = c("Galeota", "Chaguaramas", "La Brea", "Toco"), 
                     distance = c(95, 80, 90, 85), 
                     fx_cost = c(900, 670, 587, 353),
                     stringsAsFactors = FALSE)

Costs associate with this phase are:

Inbound shipping cost, based on distance or some other formula

1. Fixed Costs
2. Set up Costs
3. Variable Costs

[1] Mixed Integer Linear Programming