CMV Safety | Freight Logistics & Policy | Truck Size & Weight | TACT | Wireless, Vehicle-to-Infrastructure Communication

Hughes, Feb 2010 (Input to TRBVIS NING site)

RE: Data Visualization for Freight and Logistics

A great deal of interest is currently being shown in the potential value of visualization in a number of different transportation areas. One such example is the area of 'freight.' The reader is referred to the current TRB Research Need Statement described online at: http://rns.trb.org/printview.asp?ids=23011|. The need to utilize visualization to better understand freight data represents a good example of what is referred to as 'data visualization.' As opposed to the prevailing interest in the transportation community of using visualization to represent the physical 'appearances' of a particular project (in most cases a typical 'facility' such as a roadway, bridge, etc.), data visualization seeks to use graphics in ways that increase the ability of the user of such data to more intuitively understand key relationships between critical system variables. In this sense, traditional applications of Geographic Information Systems (GIS) are more closely related to the intent of data visualization than what currently characterizes visualization in the transportation field.

Let's turn our attention to the domain of 'freight' and what perhaps is the motivation of wanting to visualize freight data. If one looks at what is typically considered freight 'data' (e.g., the type of data one finds in the FAF2 or data in the Commodity Flow Survey from FHWA) one finds data that, for the most part, are descriptive of the 'attributes' of the system and not the system dynamics of how goods are actually transported. One finds data on the number of tons transported by different modes (truck, air, ship, etc.), the miles associated with each of these modes, the range and distribution of commodity types transported, the estimated 'value' of each class of, etc.

One is hard pressed however to discern from such data the dynamic aspects of freight system operations (by what routes are these commodities transported, what are the typical travel/transport times associated with their movement through the system, when do these movements take place, etc. All of these parameters lend themselves to be visualized in a number of ways; most notably by simple graphics (as opposed to 'tables'). The closest that the 'highway' community presently comes to understanding the operational needs of the freight community is in its focus on 'congestion' and the negative impacts of congestion on transport/delivery times, as well as the impact on air quality, especially in large urban area. We need to consider the possibility that what people are asking of visualization in the area of freight is not an ability to visualize the physical attributes of system operation (although there are times when such an emphasis is perfectly appropriate - as in the above cases), but those variables of which efficient system operation is a function. Perhaps what we should be focusing on is what is commonly the domain of 'logistics' and not freight, per se.

An analogy might be an effort to understand the dynamic operation of 'traffic' from a study of the more basic design elements of the roadway. While the latter provides a useful description of the physical attributes of the system, it fails to provide an understanding of the variables of which traffic 'operations' are a function. The visualization research need statement put forth by the TRB visualization committee (ABJ95) and described online at: http://rns.trb.org/dproject.asp?n=13836 is an attempt to draw attention to the need not only to model 'structure' but the need to capture through modeling and simulation the dynamic relationships between the variables that describe the structure in engineering terms.

A good example of where this is beginning to take place is the use of travel time data (in part collected by the highway department from a variety of sensor types) for route planning and scheduling of commercial shipments. While this represents a case where operational system attributes are being captured and displayed for public/private use, we have yet to exploit the full potential of using such data (through real time modeling and simulation) to bring about an order of magnitude improvement in operations.

Considering then the possibility that those interesting in the visualization of freight data are more interested in gaining a better understanding of the more dynamic nature of the freight system than its system outputs or attributes, what might our approach to this research need statement be? I would suggest that a starting point would be to ask, 'for what purpose does the stakeholder want to visualize freight data?' Is it to identify local, state, regional, or national level trends in the range and distribution of commodities being transported, now or into the future? Is it to identify emerging or changing routes by which different commodities are being transported? Is it to explore alternative or most cost effective modes of transport? Ultimately it is to ensure the quickest, cheapest, and most safe way to get from point A to point B.

We know that there is great interest in those who operate seaport facilities on the East Coast of the United States in projections of the future impacts of post Panamax container vessels traveling through the widened Panama Canal to the ports on the East Coast. Projections are that there will be a significant increase in container traffic (of which each of these facilities wants to capture its fair share), that the draft of these larger vessels will require harbors deeper than those that are currently in place, that additional berths and off loading equipment will be required, as well other improvements to port side facilities. There are those on the West Coast who, at the same time, are looking at these impacts and considering alternatives such as an intermodal (rail) land bridge to Midwest destinations. There is also interest within the trucking community (which is said to transport upward of 80 percent of all freight within the US) in whether there will be sufficient infrastructure (roads, etc.) to support what is projected to a doubling of freight by weight between now and 2020. In fact, the recently released FHWA 2009 Freight Facts and Figures document provide data which suggest that the rate of increase in the tonnage carried by trucks in this country between now and 2050 will increase at a significantly faster rate than the rate of growth of the general population. To meet these demands, the trucking industry is lobbying Congress to permit significantly longer and heavier combination (more productive, in their terms) vehicles in order to 'stay competitive.' The impact of such longer, heavier combination vehicles on safety and infrastructure maintenance remain unclear. While the trucking industry may become 'more competitive' there is a yet-to-be determined 'system cost' which must be borne by the country as a whole.

Whether you manage a port facility, run a trucking company, operate a state DOT, provide the services of a shipper/broker, etc., freight data and freight data projections are important - both data that describe the more physical attributes of freight as well as data that describe current and future operations. When all is said and done, ones primary concern with the movement of a commodity is how quickly, how cheaply, and how safely it can be moved from point A to point B. In a global supply chain economy we are recognizing that the freight and logistics focus has to be multi-modal and inter-modal. While there will always be 'facility' concerns (location, capacity, etc.), we are recognizing that competitiveness will require an increasing focus on 'operations' (i.e., logistics). In that sense, our focus on the application of data visualization to freight must also include improved data visualization of system operations and logistics. If the solution to the data complexity problem were obvious there would not be the current research need statements in these areas.

You can't understand traffic flow by visualizing the depth of the concrete