Navigation Investment Model

Oak Ridge National Laboratory
U.S. Army Corps of Engineers, Planning Center of Expertise for Inland Navigation
Goals and Objectives: 

The goal of the Navigation Investment Model (NIM) is to support development of an investment plan for a given subset of the inland waterway system that will maximize the net benefit to the country over a long term planning horizon (up to 70 years).


A blending of simulation, optimization and partial equilibrium models provide a flexible set of tools for planners.  In particular, the system employs:

  • Event trees and hazard functions to model lock failures, closures and repairs,
  • Demand curves by origin, destination and commodity to define shipping forecasts,
  • Optimal configuration of tows for each movement including selection of towboat, number of barges, and reconfiguration by river segment to minimize cost,
  • Congestion-based costs on a lock-by-lock basis and annual system-wide traffic equilibriums,
  • Automated calibration of the model to historic traffic using optimization routines,
  • Optimal selection and time sequencing of alternatives for capacity and reliability improvement, and
  • Determination of optimal lockage fees (optional).
Overview and Description: 

Waterways are a very attractive alternative to rail and highway transportation of heavy bulk cargo such as coal, grains, and building materials as long as the travel times are within reasonable ranges.  However, as the travel times increase due to congestion at the locks, additional operating costs for towboats and barges begin to cut into savings.  Lock congestion is created by high demand, limited capacity, maintenance closures, and construction closures.  Eventually, delayed shippers will respond by decreasing the amount shipped by water, either shifting to another mode or reducing total demand.  Investments in the waterway system can increase capacity and decrease unplanned closures. The goal of the Navigation Investment Model (NIM) is to support development of an investment path that, over a long term planning horizon (up to 70 years), will maximize the net benefit to the country. 

ORNIM supports the analyst modeling the movements on the river with modules in three areas:

  1. Analyzing lock reliability and component failures.
  2. Analyzing demand projections vs. system capacity to determine equilibrium waterway traffic.
  3. Selecting and sequencing replacement, repair or modernization efforts over the planning horizon. 

The NIM system is composed of three functional modules that interact to provide the analytical capabilities described above.  These three modules are the Lock Risk Module (LRM), the Waterway Supply and Demand Module (WSDM), and the Optimization Module.  

While the functional connectivity diagram implies that the three software components of NIM—LRM, WSDM, and Optimization—are connected to each other, these components actually exchange data through a centralized relational database.  Each of the components can be run independently, extracting data from the database and updating the database with results.  The typical sequence is to run the Lock Risk Module for each of the locks, then run the Waterway Supply and Demand Model for all potential planned closures (construction and major renovation alternatives), and then run the Optimization Module.  ORNIM has been designed to use Microsoft’s SQL Server for the database management system.

Spatial Detail: 

NIM operates on the entire inland waterway system or a subset of the system as defined by the user.  The waterway network is defined in the database.  The current default system includes 12,000 miles of waterways including the Gulf Intercoastal Waterway.  Demand and traffic are modeled at the annual level between “ports”.  Typically, a collection of docks in a region are modeled as a port.   

Mode Detail: 

NIM only models waterway traffic.  The shipment configuration details for each movement (origin/destination/commodity/barge type combination) are determined by a dynamic programming based algorithm that optimizes the tow configuration (towboat and number of barges per tow) for each segment of the river.  The goal of the detailed tow configurations is to create a refleeting plan with miminal cost conforming to the size limits by river segment.


Commodity Detail: 

Commodities can be defined at the user’s required level of detail, but it is typically defined at either the two or four digit level commodity code, using Waterborne Commerce commodity codes.  Each movement is defined by a demand curve specifying tons of demand for movement of the commodity from an origin to a destination port by a given barge type at various levels of cost per ton.  


NIM supports the analyst in three primary ways.  The first is the estimation of the future barge traffic on the inland waterway system of interest.  These estimates of river traffic are of interest to many groups including the shipping industry, waterway users, and environmental organizations.  The estimates provide the foundation for estimating the savings incurred by the use of waterway transportation vs. alternative land-based transport.    The traffic levels also provide the major input for environmental models that predict the environmental effects and mitigation required. 

The second use of the system is to estimate the long-range costs of operating the river system based on different potential maintenance plans.  Given the failure patterns of lock and dam components over time, the cost of repairs and replacements, and a maintenance protocol, the system can estimate the cost and efficacy of various maintenance plans.  This cost is broken into two components—the repair cost and the cost to the shipping industry due to delay during planned or unplanned closures.  This ability to quantify the cost impact of individual components on the system is essential in a system where maintenance funds are limited.

The third major use of the system is the evaluation of multiple alternatives.  Given options for expending funds and improving the engineering efficiency of locks or the reliability of components, the system selects an optimal set of investments and the implementation timing for those investments.  This capability allows the analyst to consider numerous combinations of measures simultaneously.  Although limited to considering smaller sets of alternatives by current processing and algorithmic limitations, the automated tools help planners build lock level plans and then integrate them into system plans. 

System Platform: 

NIM operates in a Windows environment using SQL Server for the database.

Data Sources: 

NIM does not connect live to other data sources.  Input data is developed from sources such as the Lock Performance Monitoring System (LPMS) and the Waterborne Commerce data system.  Lock transit curves relating average transit time to tons per year are developed using the Waterway Analysis Model (WAM), a lock simulation system.  Transit curves can also be developed using statistical procedures.  The network data is based on the waterway network found on the Navigation Data Center’s website ; however, additional manual editing has been undertaken to refine the network.  Lock performance data and historical traffic levels are also available at the website.

The engineering team for a study will need to supply lock component failure probabilities over time in the form of hazard functions, repair and replacement event trees including costs of repairs and closures during the repair time frame.

Example Inputs & Outputs: 
  • Inputs:  Lock component failure probabilities and repair event trees; demand curves for forecasted freight transport demand and lock transit curves; alternatives for investment with costs, closures and new transit curves. 
  • Outputs: Lock component failure probabilities by year.  Projected traffic by lock by year.  Delays and costs for commodity movements.  Optimal investments and sequencing.
Software Information: 

The NIM software (Version 5) evolved as a national model from an earlier version known as the Ohio River Navigation Investment Model (ORNIM).  NIM has been through the USACE certification process and was approved as a corporate model on February 16, 2012.  Currently, the software is only being used within the Corps of Engineers.

User Manual/User Interface: 

A user manual is in draft form. The user interface provides tools to run the model and view some of the output in summary form. Deeper analytics usually require exporting the detailed results from the database into other analytical tools.

Website & Contact Information: 

Planning Center of Expertise for Inland Navigation (513) 684-2612