Classifications of Energy Freight-Transportation Networks
The energy-transportation process is usually divided into three parts, two of them being similar. The process begins at the production point, which can be a port, a terminal, a petroleum platform in the middle of the ocean, or an electricity production zone, and generally any production point for any kind of energy. The process next comes to containers of energy. Energy is mainly transported through railways, shipping lines, and energy container trucks. It can be said that pipelines and cable networks are other examples of energy transportation. However, these are the facilities mostly used to distribute the energy rather than transport it. Transporting energy usually includes long distances and large amounts that pipelines and cable networks are not capable of handling. The third and last part of the process is the receiving ports or terminals, where the energy goes through distribution to be delivered to the consumption points.
Freight transportation also can be categorized into shippers, carriers, and governments. Shippers originate the demand for transportation, whereas carriers are utilities such as railways, motor carriers, and shipping lines. Governments are responsible for providing transportation infrastructures such as railroads, ports, and Platforms and to pass relevant laws.
Another classification of energy freight-transportation components, like other types of transportation, has two main elements: demand and supply [15].
| Logistics Operations and ManagejriL |
| fvtcKleling the Energy Freight-Transportation Network |
With respect to distance, freight transportation can be divided into long-haul versus short-haul transportation. In long-haul freight transportation, the carrier moves over long distances and between national or international ports and term- ,nals. Large vessels, railways, and, in rare cases, trucks are used in this type of transportation. However, in short-haul transportation, energy and related products are transported by large trucks. Although trucks are the main means of short-haul
transportation, railways are also appropriate when there is a need to deliver large amounts of energy—for instance, gasoline—to short distances.
Clearly, government plays a large role in developing institutions responsible l'or facilitating the transportation process. Governments contribute the infrastructure- roads, highways, and a significant portion of ports, internal navigation, and rail facilities. Governments also regulate (e.g., dangerous and toxic goods transportation) and tax the industry [14].
The following sections are brief notes about the different components of the transportation process.
Energy Production and Receiving Point
The process of energy freight transportation begins at a production point such as an oil platform. Where the energy is produced and is ready to be transported, based on the type of energy, the production process and method of storage are varied. For instance, at an offshore oil platform, oil is extracted from wells and stored in industrial facilities called oil depots. Oil that is stored in depots is in the final step of refining and therefore is ready for customer use. Oil depots usually have particular reserve tanks that are used for discharging the product to transportation vehicles.
Energy Containers
Because the production points of energy freights are most often long distances from consumption points, transportation is inevitable, so selecting the suitable mode of transporting energy products is an important concern.
For transporting energy freight to customers, a number of carriers and methods can be used. The dominant ones are trucks, pipelines, marine lines, and railroads. Each method has its own advantages and disadvantages regarding issues such as reliability, cost, safety, security, and accessibility. Nowadays with the changes in environmental conditions, other elements should be taken into consideration, such as pollution problems, noise production, traffic jams, and energy consumption ol a node [16]. Choosing the proper container depends on optimizing these various criteria, and sometimes it is more beneficial to use a combination of them to better serve customers. This is called intermodal transportation.
The following part of this section introduces some common carriers for transporting energy freight.
Railways
Rail is one common method of freight transportation. This is a cost-effective method, especially for carrying energy freights. Although this method has less speed and somehow lower reliability, it costs much less than other methods, thus making freight more affordable. Moreover, compared to truck transportation, it can transport bulkier and heavier commodities such as coal, chemicals, and petroleum in large volume to more distant areas [16]. In the United States, coal is the leading commodity of rail transportation. Another advantage of railroads is that service providers can use existing infrastructures; in most countries, governments provide the infrastructure and therefore it needs less investment [14]. However, in some countries, especially underdeveloped ones, not all of a region is covered by railways. As a result, there is less opportunity to use this mode to transport energy freight on a national scale.
Shipping Lines
Among the different modes of transporting energy freights, maritime transportation is most often used to transport critical and strategic energy commodities such as oil and related products between countries and continents. Today, more than 60% of all oil is transported by ships. Maritime lines are used to transport crude oil and its products, and their related costs are often less than other modes.
Special oil tankers are used to transport oil; these are the largest vessels in the world [17].
Trucks
Trucks are among the most popular methods of transporting commodities within and between countries. It is the leading way to transport commodities in most countries such as the United States. Like other modes of transportation, it has advantages and disadvantages. Its wide coverage, convenient accessibility, fast responsiveness, and flexibility make it popular. Furthermore, in some circumstances the application of trucks and road transportation is inevitable, because not all consumption points are connected by rail or water. Although in most cases it is more suitable for short distances and lightweight shipments, it is more costly than rail transportation.
In addition to that, this mode of transportation has disadvantages that must be considered in making decisions, such as its pollution and the amount of fossil fuels it consumes. It also causes traffic congestion and has low levels of safety. In most countries, the number of road accidents and tolls are very high, and it is not possible to get rid of it completely even with precautionary rules and standards [16].
Demand and Supply
Travel demand derives from the need for energy in other parts of the region or the world that is deprived of energy resources. As shown in Figure 21.4, the demand flows in transportation systems rise from the fact that there are discrepancies in each and every freight transportation—that is, the distance between production and receiving points varies from one place to another, the energy containers stretch from trucks to vessels, and the types of energy vary from petroleum to NG. Their movements make up freight travel's demand flows.
21-2.4 Introducing the Energy Freight-Transportation Network Models
The network of energy freight transportation is of different levels, from regional movements through highways and trucks, to national movements through railways, and to international movements through shipping lines and large vessels. Designing
Supply
| Transportation facilities and services |
| Transportation service performance * |
Supply element capacities
Congestion
Flows on modal networks
Travel demand by ^
| Transportation system |
transportation mode
|
Level and spatial distribution of travel demand
Demand
Figure 21.4 Relationships between the transportation system and the activity system [15J.
and evaluating the models of such networks requires quantifying interactions among the elements of existing and future transportation systems [15].
Although every element cannot be identified or controlled in modeling, it still plays a central role in the design and evaluation of transportation systems. Factors such as a region's or country's transportation infrastructure, constraints of delivery points, and marine and road traffics all influence the behavior of a network model but can hardly be modeled or controlled. It is therefore necessary for scientists of energy-transportation networks to account for these hidden parameters when designing a network model.
Transportation planning, from goods to energy transportation, has been widely discussed in books and papers, but most of them are about road transportation by truck rather than other modes of transportation. It may be questioned why there is a
lower level of attention, in spite of the large capital investments and operating costs associated with these other modes. Although research on rail planning problems has increased considerably over the last 15 years, it is not the same for maritime
transportation.
Christiansen [18] has some explanations. First, there is low visibility; people mostly see trucks or trains rather than ships, and ships are not the major transportation mode worldwide. In addition, large organizations that sponsor research mostly operate fleets of trucks, not ships. Second, the planning problems of ship- pine networks are less structured than the other modes. This makes the planning more expensive because of the customization of decision-support systems. There is more uncertainty in maritime operations because of weather conditions, mechanical problems, and incidents such as strikes. Slacks in maritime transportation planning are few because they have high costs. Most quantitative models originated in vertically integrated organizations where ocean shipping is just one component of the business. This occurs because there are many small family-owned companies, because the ocean shipping industry has a long tradition and it is not open to new ideas.
Modeling the Transportation of Hazardous Materials
The US Department of Transportation (USDOT) defines a hazardous material as any substance or material capable of causing harm to people, property, or the environment [19]. It has categorized a list of hazardous materials into nine classes according to their physical, chemical, and nuclear properties. Gases and flammable and combustible liquids are among the classes.
It should be mentioned that most hazardous materials (hazmats) originate at locations other than their destination. Oil, for instance, is extracted from oil fields and shipped to a refinery (typically via pipeline); many oil products, such as heating oil and gasoline, are refined at a refinery and then shipped to storage tanks at different locations within a country or abroad.
The risks associated with the transportation of oil and gases and their consequences can be significant because of the nature of the cargo: fatalities, injuries, evacuation, property damage, environmental degradation, and traffic disruption. Reductions in hazmat transportation risks can be achieved in many different ways. Some of these ways are not related to modeling and planning the transporta- llon network, such as driver training and regular vehicle maintenance. Others can be studied through operation research and modeling.
As mentioned in previous sections, energy can be moved over roads, rails, or Water. In some cases, shipments are intermodal; they are switched from one mode to another during transit. Hazmat transportation incidents can occur at three points: the 0rigin when loading, the destination when unloading, and en route. To identify |he route that minimizes fuel costs and travel times between production and receiv- ln§ points, operation research models are designed with the related constraints.
According to different routes, energy transportation as a kind of hazardous mate- rial is a typical multiobjective problem with multiple stakeholders that are difficult to solve. Transport by truck, for instance, has choices between selecting short routes while moving through heavily populated areas or selecting longer routes through less populated areas, which makes the transportation cost more and expose to risks
Mathematical models that are described in the following sections allow representation and analysis of the interactions among the various elements of a transportation system.
Components of Energy Freight-Transportation Models
Modeling any transportation network requires identification of components that are acting reciprocally. Ghiani et al. [20] introduce cost as the major component of the transportation model and then classify the problems based on relevant costs.
As mentioned in the previous section, despite the fact that factors affect the transportation network model, they can hardly be identified, quantified, and modeled. Some of the main factors are categorized under the name of external factors, which is a subcategory of operational factors [21].
The transport infrastructure is of great importance. Lacking such a capability could affect the scheduling and delivery of the energy shipment. In some regions, there are not proper rail networks and essential facilities in the terminals to transfer energy to a location. Ports have to be well equipped for large vessels to berth and transfer the energy freight.
In addition to that, a transportation network is affected by trade barriers as well as laws and taxation policies. Variation in any of these parameters around the world may affect the decision concerning the most appropriate mode of transportation and routings for cost reasons. Legal requirements are likely to differ from one country to another. As a result, there would be problems in costs and planning while trying to adapt to the requirements.
Because parameters and problems of modeling ship fleets are different from those of other modes of transportation, ships operate under different conditions. Table 21.2 provides a comparison of the operational characteristics of the different freight-transportation modes. Shipping lines are mostly in international territories, which means they are crossing multiple national jurisdictions. In energy freight transportation with ships, each unit represents a large capital investment that translates into a high daily cost because they must pay port fees and operate in international routes.
In addition to that, other means of energy freight transportation generally come in a small number of sizes and similar models and designs, whereas among ships we find a large variety of designs that result in nonhomogeneous fleets.
More than that, ships have higher risks and lower certainty in their operations because of their higher dependence on weather conditions and on technology, and because they usually pass multiple jurisdictions. However, because ships operate around the clock, their schedules usually do not have buffers of planned idleness that can absorb delays. As far as trains are concerned, they have their own dedicated rights of way, they cannot pass each other except for specific locations, and their size and composition are flexible (both numbers of cars and numbers of power
Table 21.2 Comparison of Operational Characteristics of Freight Transportation Modes [18] Operational Mode
Characteristics----------------------------------------------------------------------------------------------
Ship Aircraft Truck Train Pipeline
|
| NA. not applicable. |
units). As a result, the operational environment of ships is different from other modes of freight transportation, and they have different fleet-planning problems.
Energy Freight-Transportation Costs
There are different costs during a transportation network. They can be divided into transportation costs and handling costs [15]. Transportation costs include the cost of operating a fleet, the cost of transporting a shipment, the cost of hiring carrier if not owned, and the cost of a shipment when a public carrier is used. Handling costs I are not discussed in energy freight transportation, because they are incurred when inserting individual items into a bin, loading the bin onto an outbound carrier, and reversing these operations at a destination.
The Cost of Operating a Fleet
The main costs are related to crews' wages, fuel consumption, container depreciation, maintenance, insurance, administration, and occupancy. It is obvious that wages and insurance are time dependent, fuel consumption and maintenance are distance dependent, and that depreciation depends on both time and distance whereas administration and occupancy costs are customarily allocated as a fixed annual charge.
The Cost for Transporting a Shipment
This type of cost is paid by a carrier for transporting a shipment. It is rather arbitrary because it would be difficult to assign a trip cost to each shipment, where several shipments are moved jointly by the same carrier—that is, a large vessel containing barrels of petroleum and other downstream products simultaneously.
The Cost of Hiring Carrier
Although hire charges are parts of a transportation total cost, they are still unidentified and hard to evaluate.
The Cost of a Shipment Using a Public Carrier
The cost for transporting a shipment when using a public carrier can be calculated on the basis of the rates published by the carrier. The size and equipment of a carrier as well as the origin, destination, and route of the movement are factors that are taken into account when calculating this cost.
Risk
As discussed in "Modeling the Transportation of Hazardous Materials" section, energy in the form of gas and oil is one type of hazardous material. As a result, possible incidents during loading, transporting, and unloading should be considered when making models. To estimate the probability and cost of a hazmat release incident, various consequences must be considered. The consequences can be categorized as injuries and fatalities (often referred to as population exposure) |22,23], cleanup costs, property damage, evacuation, product loss, traffic incident delays, and environmental damage. It is clear that all impacts must be converted to the same unit (e.g., dollars) while modeling in order to permit comparison and complication of the total impact cost.
Route
Some models presented in the field of energy-transportation networks seek to minimize travel distances between production and consumption points. It first occurs that the shortest possible route—roads and railways or marine lines—would be the answer. However, looking profoundly at all of the issues concerning routing problems shows that there are significant components that prevent the model from being designed and solved in such an easy way.
The previous sections contain explanations about the parameters dealing with routing problems. As mentioned, not all shortest distances have the lowest expense. Models of freight transportation seek to solve a multiobjective function in which more than two factors are optimized. A routing model should give decision makers the shortest route with the minimum cost simultaneously. Because it would be quite hard to achieve such a solution, the models show an appropriate solution that does not necessarily have the minimum distance or cost.
More than that, previous sections explained one important issue that has arisen in recent years. The security of the routes matters considerably as the rates of lost or attacked energy freight increase. There are routes with lower levels of security that have a minimum cost or distance. Meanwhile, secure roads or marine lines certainly cost more for longer distances. Routing model planners have to design models that can achieve a good solution while at the same time accounting for as many issues involved in the problem as possible.
Models of Energy Freight-Transportation Network
Modeling problems of energy freight-transportation networks contain assumptions, constraints, and one or more objective functions. Models usually focus on one attribute of the network—for instance, minimizing the cost of moving energy while ignoring other effective attributes or considering them as constant parameters.
As discussed in previous sections, particularly "Energy Containers" section, modes of energy freight transportation vary from trucks to trains to fleet. The tactical planning level perspective is missing in ship routing and scheduling studies reported in the literature. Fleet scheduling is often performed under tight constraints. Flexibility in cargo quantities and delivery time is often not permitted. So the shipping company tries to find an optimal fleet schedule based on such constraints while trying to meet the objective functions—that is, maximizing profit or minimizing costs. Br0nmo et al. |24] and Fagerholt [25] have developed models that consider flexibility in shipment sizes and lime windows. The models are not specified in energy but would be applicable in shipping energy problems as well. The results of their studies show that there might be a great potential in collaboration and integration along the factors of a transportation process—for instance, between shippers and shipping companies.
Christiansen et al. [18] introduce a planning problem in which a single product ,s transported and call it the single-product-inventory ship-routing problem (s-ISRP). The assumptions and constraints of the model are close to reality—that ls> transporting energy using ships. The production and consumption rate of the transported product—energy, in this case—is constant during the planning horizon. The advantage of the model is that contrary to similar scheduling problems, neither the number of calls at a given port during the planning horizon nor the quantity to be loaded or unloaded in each port call are not predetermined. There needs to be some initial input in order to determine the number of possible calls at each p0rt the time windows for the start of loading, and the range of feasible loads for each port of call. The initial information would be the location of loading and unloading ports, supply and demand rates, and inventory information at each port. Eventually the planning problem finds routes and schedules that minimize the transportation cost without interrupting the production or consumption processes.
Ghiani et al. |20] continue the problems based on transportation cost, discussin» freight-traffic assignment problems and classifying them as static or dynamic Static models are appropriate when decisions related to transportation are not affected explicitly by time. The graph G = (V, A) is then applied, where the vertex set V often corresponds to a set of facilities as terminals, ports, and platforms in production and receiving points, and the arcs in the set A represent transportation carriers linking the facilities.
In addition to that, they take a time dimension into account in dynamic models, including a time-expanded directed graph. In a time-expanded directed graph, a given planning horizon is divided into a number of time periods, Tl, T2..., and a physical network is replicated in each time period. Then temporal links are added. A temporal link connects two representations of the same terminal at two different periods of time. They may describe a transportation service or the energy freight waiting to be loaded onto an incoming carrier.
Some linear and nonlinear models based on cost parameter are as follows: minimum-cost How formulation; linear single-commodity, minimum-cost flow problems; and linear multicommodity, minimum-cost flow problems.
As explained in "Modeling the Transportation of Hazardous Materials" section about oil and gases as types of hazardous materials, transporting them contains risks that have to be measured. Erkut et al. [26] talk about risk along an edge or route while transporting hazmats in what they call linear risk. They focus on hazmat transportation on both roads and railways. A road or rail network is defined as nodes and edges. The nodes stand for the production and consumption points, road or rail intersections, and population centers. The road segments connecting two nodes are called the edges. It is assumed that each point on an edge has the same incident probability and level of consequence. As a result, a long stretch of a highway or railway moving through a series of population centers and farmland should not be represented as a single edge but as a series of edges. This is the difference between a hazmat transportation network and other material networks. Erkut and Verter |27] discuss this difference as a limit to the portability of network databases between different transport applications. Also, along with Erkut and Verter [27], Jin et al. [28] and Jin and Batta [29] suggest a risk model that considers the dependency to the impedances of preceding road segments.
Transporting energy from place to place requires a detailed plan and a schedule in order to minimize the costs during the process and determine the shortest route in time windows while accounting for the probability of incidents. This fact makes researchers model the realities and propose varieties of models to solve the problems. The models cover different transport modes. Erkut et al. [26] have provided a classification of papers reviewing different problems. Table 21.3 presents an extended version of what they have done. Not all of the research shown in the table concentrates on energy transportation, but some of it discusses models of transporting hazmats such as energy.
Some research has also focused on designing a transportation network. The networks are used to transporting hazardous materials in general, but they may also be applicable for energy freight transportation. Some of them are as follows: Berman et al. [65]; Erkut and Alp [66]; Erkut and Gzara [67]; Erkut and Ingolfsson [39]; Kara and Verter [68]; and Verter and Kara [69j.
Although the cost of a transportation network is a significant factor, other parameters also act on the network. A transportation network service problem which is in the operational level consists of deciding on some elements. The elements include the characteristics (frequency, number of intermediate stops, etc.) of the routes to be operated, the traffic assignment along these routes, and the operating rules and laws at each terminal [20].
Service network design models can be classified into frequency-based and dynamic models. Variables in frequency-based models express how often each transportation service is operated in a given time horizon, while in dynamic models a time-expanded network is used to provide a more detailed description of the network. Models of service network design in both categories are fixed-charge network design models, the linear fixed-charge network design model, the weak and strong continuous relaxation.