TopTank

Dynamic simulation
Objectives
Terminal Infrastructure
Input
Results

Dynamic simulation

A generic model of a terminal has been built with open source software Python using the Salabim module. Salabim is used to model and simulate all kinds of discrete processes. As the objects defined within the model interact dynamically, the simulation reflects reality very well. Entities arrive in queues and waiting times are recorded.
Within the generic model for terminals vessels (barges, trains and trucks) arrive according to a predefined schedule or stochastic arrival process. These vessels occupy equipment (tanks, jetties, jettylines, connection lines, headers and pumps) that are necessary to fulfil their objective: loading or discharging a certain amount of product to a certain tank of a certain customer. When occupied, this equipment can not be used by other vessels, thus creating queues and waiting times. Occupancy rates of equipment are recorded.


Objectives

The objectives of using a simulation model for terminals are:

  • Improve the design of new terminals:
    • Find bottlenecks
    • Test the impact of adding / removing equipment
    • Test the impact of dedicated versus flexible systems
    • Change product allocation
  • Test the impact of changes in existing terminals, for example:
    • Increase in throughput
    • Changes in infrastructure: adding/removing equipment, changing connections
    • Change planning from FIFO to timeslots
    • Changes in vesselsizes, parcelsizes, frequencies
    • Changes in product/customer portfolio
  • Test future contract details with customers
  • Test future service levels with customers
  • Convince stakeholders: statements on logistics of the terminal are supported by results of the model.


Terminal infrastructure

With the aid of tables the complete terminal infrastructure is recorded in the model: tanks, berths, jetty lines, pumping platforms, pumps, pumpheaders, headers and connection lines. The model contains a navigation system that determines all possible routes between tanks and berths and puts them in a table. When a vessel arrives the most optimal available route for pumping the product is chosen from this table. All equipment that is part of this route is occupied when the pumping starts. When putting together the route table the navigation system takes into account constraints like a maximum number of pieces of equipment, use only equipment of the same product group and use a suction and a pressure connection. The simulation model reads this route table before a run is started.


Input

A model of a terminal is constructed by filling the generic input tables on the website or uploading .csv files. The TopTank simulator has been built in Python and consists of the infrastructure of the terminal, the generation of Modes of Transport (MOT can be ships, trains, trucks and pipelines) and a planning module.

Settings

  • Transfers: from MOT to tank, from tank to MOT, from tank to tank and from MOT to MOT
  • Planning: MOT will search for the combination of line-ups with the shortest total turn-around time. MOT in the queue will make equipment reservations in order of arrival. Ships may wait at anchorage for all line-ups for all the parcels to be available or only for the line-up of the first parcel. Planning will select the first MOT in the queue where all lights are green. Each MOT type has its own queue. Priorities per MOT type can be set
  • Definition of product types and categories. Equipment defined in a certain product category can transfer different product types. Setting a change over time per product category determines the time that it takes to change from one product to the other within this product category

Infrastructure

From input tables the simulator will construct all possible line-ups on the terminal. The input consists of the following tables:

  • Definition of berths: number of simultaneous loading arms, blocking of other berths, priority, capacity in relation to dimension of MOT types
  • Definition tables for equipment: tanks, berths, pumps, headers, jettylines, connection lines manifolds. In the definition tables, the following parameters are defined per equipment: capacity, flowrate, product (type or category), change over time
  • Construction tables for line-ups: types of line-ups, connections between equipment

MOT generation

MOT can be generated by using historical data, by random generation and by a combination of both. The input consists of the following tables:

  • Vessel classes: dimensions (determined by the capacity of berths), pumping rate, priority (for planning module), idle times and transit times
  • Historical data with per MOT arrival: vessel class, arrival time, im-/export. Each arrival may carry up to 10 parcels. For each parcel product type, size and tank number should be given
  • Random generation: per vessel class, product type, im-/export, cargo size, number of MOT per month (for trucks an hourly frequency per day can be determined. (Tank to tank transfers are defined in a separate table). The Generic Tank Terminal Simulator, as described above, has been used in a wide variety of tank terminals. We are convinced that all thinkable scenarios of your terminal can be modelled by changing the input in Excel tables


If interested in using the model,
please send me an email


Results

The model output consists of logfiles with timestamps generated per single MOT and equipment. From these logfiles all possible management information can be created. The usual KPI’s in the dashboard are:

  • Throughputs per product, MOT type
  • Volume in tanks over time
  • Berth/station occupation: specified to any MOT attribute: MOT type, productgroup, customer, etc.
  • Waiting times:
    • Per berth/loading station, product, MOT type (and combinations
    • Distributed to berth, infrastructure or planning
    • Averages, distributions
  • Queue lengths over time
  • Turnaround times of MOTs: same specification as waiting time
  • Occupancy rates of berths/loading stations and equipment divided over berths and/or products

Especially the waiting time for infrastructure should be paid attention to as this time can mostly be reduced by simple and inexpensive measures. In most times investment in an extra jetty can be postponed.