Freight Production Research Articles

Modelling urban freight generation using linear regression and proportional odds logit models

Receivers and intermediary establishments typical to urban areas have unique freight behaviour and are understudied, particularly in developing countries. This paper uses establishment-based freight survey data to examine their freight generation patterns in the Chennai metropolitan area, India. Three sets of Simple Linear Regression (SLR) models, three single predictor proportional odds logit models, one multiple linear regression model, and one multiple predictor proportional odds logit model with business size (employment, area, and operational age) and indicators of establishment category as regressors are developed each for both freight production and freight attraction. Partial proportional odds logit models are developed in a few cases to overcome the limitations of proportional odds logit models. The best SLR model varied with the establishment category for freight production and attraction. The establishment area model is the best among single predictor proportional odds logit models. The multiple predictor proportional odds logit models marginally improved the fit over single predictor models. The proportional odds logit model results show that the establishment category has a greater impact on freight generation levels than business size variables. Since earlier studies rarely focused on receivers and intermediary establishments, policymakers may benefit from the developed models and study insights while estimating freight demand and developing freight policies.

Examining the Effects of Local Sample Sizes on Spatial Transferability of Freight Production Models

Recent research in freight transportation planning has been exploring the spatial transferability of freight demand models, which can help the planning agencies of developing economies save the cost and time incurred in freight surveys. As the demand models are time-, cost-, and data-intensive, it is prudent to analyze the effects of sample size on the transferred model in a region. The findings and inferences from such analysis will save resources in freight data collection programs. Earlier, conventional models like ordinary least squares (OLS) regression were assessed for transferability. However, the predictive ability and transferability of such non-conventional models are not well studied. It is necessary to understand whether the extent of transferability of non-conventional models is greater than that of conventional models so that planning agencies can adopt more reliable modeling approaches. This paper investigates the spatial transferability of freight production models using OLS, robust regression, and multiple classification analysis (MCA). The results of the transferability assessment show that MCA models have better transferability using the naïve transfer method. In addition, transferability is assessed for different sample sizes to examine the variation in the extent of transferability. The MCA models have shown the least deviation, indicating that these models are preferred for transferability when the sample size is small.

Rail freight production in Brazil: Projecting scenarios in times of global uncertainty

The human losses caused by the COVID-19 pandemic are an overriding concern, nevertheless it is important to discuss other effects that will come in the wake of the pandemic. This article evaluates the pandemic impacts on the rail freight production in Brazil, as railways play a key role on the logistics chain of Brazilian commodities. To achieve this objective, rail production data for the period between January 2006 and December 2021 was analysed using intervention and prediction methods based on Box-Jenkins ARIMA models, assisted by Monte Carlo simulations. The study reveals that, although the 2008 economic crisis caused an inflection in the time series of the analysed macroeconomic variables, it had no statistically significant impact on rail freight production, considering the time series as a whole. The results also confirm that the historical trend of growth in production has been maintained until the moment with no apparent impact from the COVID-19 pandemic. Given the prevailing uncertainty, the study examines other scenarios, projecting different degrees of expected production instability with a prediction horizon of 2025. These scenarios could be an important aid to plan short- and medium-term actions for this transport mode that is crucial to Brazilian economy and exports.

Estimating freight production/attraction using nighttime satellite imagery

An appropriate understanding and accurate estimation of zonal freight production (FP) and attraction (FA) volume is one of the most important topics and the initial step in developing the origin and destination (OD) matrix. In general, zonal FP and FA are difficult to be obtained as observed data, particularly in developing countries, owing to the lack of large-scale freight surveys. Thus, several proxy indicators are often used for estimating FP and FA, such as the gross regional product (GRP), population, and the number of establishments. However, these indicators do not often estimate the zonal FP and FA accurately and are even not available readily. The nighttime light intensity (NLI) data that is human-induced light emission data and one of the open alternative data, possibly improves the accuracy of FP and FA estimation. However, few studies have examined the relationship between NLI and FP/FA. Therefore, the objectives of this study are to examine whether NLI can be a significant indicator for estimating zonal FP and FA relative to other socio-economic indicators and to examine the types of NLIs that can accurately estimate the FP and FA through a case study in Japan. As a result of the analysis, NLI data are identified as a significant variable for the estimation of FP and FA among various socio-economic variables. Although some of the indicators related to secondary industries provide higher accuracy, NLI exhibits better estimates than the popular indicators such as GRP and population. In addition, we propose three types of total NLI that multiply the unit NLI by the zonal area and this generates higher accuracy of unit NLI, the average light intensity of all pixels in a zone. We have also found that the use of habitable area for the estimation of total NLI provides the highest accuracy among the four types of NLIs including unit NLI. Through the analysis, NLI data is sufficiently applicable for estimating zonal FP and FA, relative to other important socio-economic indicators, although space transferability should be examined in other study areas. These findings could help estimating FP/FA particularly in countries and regions with limited statistical data available.

Freight production and attraction of industrial, agricultural and livestock, food, and fruit and vegetable commodities

Freight production and attraction of industrial, agricultural and livestock, food, and fruit and vegetable commodities

Are Freight Production Models Transferable between Urban and Suburban Areas? Guiding Model Transfer in Geographically Sprawling Indian Cities

Investigating the spatial transferability of freight generation (FG) models is an imperative research need to enable the usage of formerly estimated model parameters in new application contexts with or without the usage of local data. By understanding how to transfer models (and to what extent), planning agencies in large countries like India can save freight survey costs in regions where they lack the institutional capacity and resources. Due to geographically sprawling nature of most of the Indian cities, an important research question regarding transferability is whether the models developed for urban areas can provide accurate estimates of freight activity in the suburban areas or vice versa. This paper aims to provide two solutions to this problem: (i) compare the relative effectiveness of transferability depending up on the direction of transfer and (ii) assess which models can be transferred and which cannot. Data collected from seven cities are used for this study. A set of freight production (FP) models are developed using this data to understand the differential influence of geographical location and industry segment on the model coefficients. The estimated FP models show that suburban establishments exhibit significantly higher FP rates as compared to other establishments. Subsequently, transferability direction and accuracy are determined using standard metrics such as transfer R2, relative aggregate transfer error and transfer index. The transferability findings will provide actionable insights into development of FP models in regions with data constraints, which is of great value in an era of declining budgets for travel surveys.

Analysis on Operation Management of Logistics Park in China

Since the State Council issued the “Plan for the Adjustment and Revitalization of the Logistics Industry” in 2009, the logistics industry, as the basic and strategic industry in the development of the national economy, has played an increasingly important role in promoting industrial transformation and upgrading and promoting economic development. In order to promote the development of logistics industry, a large number of logistics parks have been built all over the country under the guidance of policies and the promotion of the market. However, after the construction of these logistics parks has been put into operation, There are a series of problems in its development and profit. This paper analyzes the present situation and profit of logistics park in China, and briefly analyzes the main business characteristics of freight hub, production service, port service, trade service and comprehensive service. On this basis, the main problems existing in the existing profit model of logistics park in China are summarized.

Freight production of agricultural commodities in India using multiple linear regression and generalized additive modelling

Freight transportation has a key role in the economic competitiveness of any nation. India is one of the fastest-growing nations in the world; its agricultural sector plays a vital role in contributing to the country's economy. In this paper, the freight production in the agricultural sector of India is modelled using multiple linear regression (MLR) and generalized additive modelling (GAM) approaches. Using district-level data, factors influencing agricultural freight production are identified and the relationship between them is modelled. The study considered 210 districts of eight states covering South India and parts of Central and Western India. Population, employment in the agricultural sector, gross cropped area, and gross irrigated area are identified as influential factors. GAM approach is a flexible method which can predict non-linear responses for the given predictor variables by allowing non-linear functions for each of them while maintaining additivity. The applicability of GAM in modelling freight production while tackling the non-linear effects of predictor variables on freight production is investigated and encouraging results are achieved. The results revealed that both MLR and GAM models have good modelling efficiency, however, the GAM model outperformed the MLR model in both fitting and predicting. The study can be used in evaluating the commodity movements and transportation demand which in turn will help in decision-making for the provision of freight transportation facilities and policymaking.

Effects of Business Age and Size on Freight Demand: Decomposition Analysis of Indian Establishments

Freight studies have traditionally used business size measures such as employment and gross floor area as predictors of freight generation, giving very limited attention to the effects of an establishment’s age on freight demand. This study uses establishment-based freight survey data collected in seven cities of Kerala, India, to analyze the impacts of the age of an establishment on its freight demand. This is achieved by grouping the establishments into pre-specified classes with relatively homogeneous freight demand pattern. This classification is based on a data-driven a posteriori segmentation of industrial classes. These groups are further divided into sub-groups based on founding year of an establishment. As the business age increases, establishments grow in business size indicators and hence the changes in productivity can be either because of age or because of size or both. It is very important to separate the effects of these two indicators. The Blinder–Oaxaca decomposition method is used in this study to disentangle the effects of size and age on freight production and freight trip production. Age is found to be a fundamental driver of freight demand in younger establishments while the explanatory power of business size variables in explaining freight demand diminishes with age. The study findings illustrate the potential omitted variable bias that can occur when freight demand is estimated using business size indicators, without controlling for the business age differences.

Evaluating interregional freight accessibility conditions through the combination of centrality and reliability measures

The main contribution of this paper is a new approach to evaluate and compare the freight accessibility conditions offered in a set of regions or cities. The key innovation of this approach lies in that it combines (statistical) centrality and reliability measures of accessibility, thus tackling one of the weaknesses that researchers have identified in the vast accessibility literature. The proposed approach is first explained in details for a hypothetical setting, and then is applied to Iran using data collected in the latest transport survey conducted in this country. The focus is placed on road transport, by far the dominant freight transport mode in Iran. The results obtained through this application provide clear insights into the relative positions of the main freight production and attraction zones of Iran with respect to road accessibility conditions, and therefore valuable information to assist transport authorities in their decisions regarding the evolution of the country's road network.

Freight generation and geographical effects: modelling freight needs of establishments in developing economies and analyzing their geographical disparities

Freight generation (FG) models are important to transportation authorities and planning agencies as they can be used to forecast local/regional/state/national freight movements for facility planning and evaluation of freight-specific investments. Compared to the modelling efforts in passenger transportation, freight transportation remains largely unexplored in developing economies like India mainly because of the absence of national wide commodity flow survey unlike in developed economies. In recognition of this, we collected establishment-based-freight-survey data from 432 establishments in Kerala, India. These data are used to classify the establishments into 13 homogenous industry sectors; several FG models are developed for these sectors. The relationships of annual freight production (FP) and attraction (FA) with establishment size variables (employment and area) were investigated with three modelling approaches. Firstly, a set of 52 practice-ready FG models are estimated using linear regression technique for establishments in each industry sector. Modelling results revealed that employment is a better representative of FP, whereas area represents FA better. The employment-based FP rates in Kerala are found to be lower than that in New York, much like what is observed in passenger transportation; passenger trip rates in developing economies are lower than that in developed countries. Secondly, FG rate tables and Nomograms are developed using Multiple Classification Analysis technique for all industry sectors considering employment and floor area as categories. These nomograms and FG rate tables may be used as planning tools for city developing agencies, while incorporating freight transportation in the overall planning process. Lastly, ANCOVA analyses is provided to assess the geographical disparities on FG and, thereby the model transferability. Study findings will be useful in developing policy guidelines for freight-specific investments, operational strategies, freight movement regulations, and taxation policies, etc. for Indian cities.

Comparative Assessment of Industrial Classification Systems for Modeling Freight Production and Freight Trip Production

Freight demand models typically employ a priori classification systems for dividing establishments into hypothetically complementary groups with homogeneous patterns in freight production (FP) and freight trip production (FTP). Although an attractive and popular notion, the assumption of homogeneity within these a priori industrial classes is reductive in nature and is not yet tested in literature. This research examines this hypothesis and explores the possibility of a data-driven segmentation by examining the relationships between FP/FTP patterns and prevalent a priori classes; subsequently, it creates homogeneous ensembles of a posteriori segments through aggregation. This research labels, explains, and interprets these novel segments using commodity value density of industrial classes. The alternate segmentation schemes are compared in their ability to predict FP and FTP and it is found that: (i) industrial classification systems (NAICS, ISIC) perform significantly better than product classification systems (ASICC); (ii) a considerable portion of variability in FTP does not depend on employment predictor due to the underlying influence of shipment size; (iii) an a posteriori segmentation scheme considering shipment size may represent an effective middle ground for developing both FP and FTP models in freight demand model systems. Adoption of these novel segments of the freight travel market has the potential to reduce the sample size requirements of freight demand model systems and minimize the financial necessities for future freight surveys.

Conceptualizing freight generation for transport and land use planning: A review and synthesis of the literature

Freight generation and movement patterns are not well understood by planners and policy-makers tasked with making complex strategic land use and transport planning decisions. In the absence of detailed planning evidence, they may rely on scant or anecdotal data, extrapolated and presented through complex quantitative models. Unfortunately, predictive model outputs can fail to accurately match observed outcomes, and such models cannot predict complex long-term phenomena which may transformatively disrupt freight production and movement patterns.Through a review of the literature, we apply Porter's Diamond Model of Competitive Advantage (Porter 1990) to develop a novel conceptual framework for freight generation. We illustrate how emerging themes and new evidence of relevant economic, environmental, social, and governmental factors can be meaningfully structured within this conceptual framework. By compiling recent evidence of spatio-temporal complexity from the literature, we highlight the utility of such a framework in assisting planners and decision-makers to incorporate a wider set of freight generation factors – particularly demand factors, land use relocation effects, shifting firm strategies, and emerging transport technologies – in the practices of understanding, modeling, planning, and managing urban freight. Application of the framework should assist in ensuring that significant factors and phenomena are not ignored in critical planning decisions, encourage the input of a more diverse set of planning expertise at the policy-making table, and throw renewed emphasis on potential qualitative and mixed-methods freight case study research.

Characterization and analysis of metropolitan freight patterns in Medellin, Colombia

PurposeThis paper seeks to pilot test a novel way to collect freight and service activity data and analyze the collected data in the metropolitan area of Medellin, Colombia.MethodsThis research collects data using a multi-layer and multi-actor approach that includes surveys to receivers, suppliers, carriers, and truck drivers. The data are used by the authors to describe the overall freight patterns in the area of study and to show lessons learned.ResultsThe data collection resulted in 2947 establishments (4.4% of the total establishments in the city), a cordon survey of 2950 commercial vehicles (17% of the total vehicle volume) accessing the urban area, and carrier interviews to ten companies and 130 truck drivers. The results indicate that a total of 33,274 metric tons/day enter the study area, 35,240 tons/day leave the area; while 7000 tons/day are distributed in the study area. In terms of freight trips, 6600 trips/day enter the study zone and 6600 trips/day leave it.ConclusionsThe data collection effort enabled the analyses of freight generation patterns. The freight surveys used in the study complement each other, and provided a good depiction of the freight movements in urban areas. It was found that in the Medellin Metropolitan Area, freight-intensive sector establishments generates, on average, significantly more cargo (freight attraction plus production) than the service-intensive sectors. The analyses of the surveys allow the decision makers to understand the nature of the cargo and the generation patterns in different type of establishments. This characterization of the freight patterns is vital for the forecasting of the behavior of the cargo and it is the main input to perform freight demand modeling for city planning, especially for developing countries, where there are too many budget constraints.

Freight Demand Forecasting Considering Economic Growth Factors

This work proposes a methodology to project future freight demand for all commodity types that begin and end in each geographical region and the amount of freight that moves between all origin and destination pairs. Following the traditional four-step demand forecasting framework, the procedure corresponds to trip generation and trip distribution analysis for interregional freight demand. Using future economic growth factors from macroeconomic and input-output models, the amounts of freight production and attraction in each analysis zone are forecasted and taken as given. Subsequently, an iterative matrix balancing method is applied to determine the estimated freight shipment demand for all origin and destination zone pairs. The proposed algorithm is applied to generate predicted future freight demand within the United States from 2010-2050 in five-year increments based on the national freight demand data from 2007. Four different scenarios are proposed that consider variations in both global economic growth and environmental regulation. This study will assist transportation planners and decision makers in public and private sectors to assess how future freight delivery demand on the national scale considering various future global economic growth and environmental policy scenarios will affect various issues such as air quality and human health problems.

Aggregate Freight Generation Modeling

The most comprehensive publicly available freight databases are the Commodity Flow Survey and FHWA's Freight Analysis Framework. These two sources contain the dollar value and weight of freight movements at high geographic levels, such as states or metropolitan areas. Because of the difficulty of obtaining freight data at lower geographic levels, various practitioners and researchers have suggested estimating freight models on the basis of aggregate data. Following recent practice, a methodology is presented to estimate a nationwide production and attraction model for U.S. domestic trade of goods. To this end, a Commodity Flow Survey data set provided by the U.S. Census Bureau and composed of a 2-nonconsecutive-year period (2002 and 2007) of goods movement between U.S. states for 27 industry sectors was used. The state payroll by industry sector, obtained from the County Business Patterns of the U.S. census, was the variable used to estimate freight generation models. The main objective of this study was to analyze the temporal stability and predictability of the proposed aggregate models. The results indicate that payroll alone explains a significant portion of the freight production and attraction at the state level. However, such simplification in the model process did not result in reasonable predictions of freight for a future-year horizon. It is recommended that time-dependent factors (e.g., variables related to changes in industry productivity) affecting freight demand be considered in the modeling process.

Time-Dependent Effects on Parameters of Freight Demand Models

Seven national freight origin-destination samples collected in Colombia from 1999 to 2005 were used to conduct an empirical investigation of the temporal stability of parameters of freight demand models in the short to medium term. Freight generation, freight distribution, and empty trip models were considered. To identify time-dependent effects, models were estimated with a panel formulation with time-dependent parameters and fixed time effects and then compared with the corresponding cross-sectional models. The results indicate the presence of statistically significant time-dependent effects on all freight generation models (production and attraction), freight distribution models (based on both loaded vehicle trips and commodity flows), and empty trip models. A literature review indicates that few studies are available on the temporal stability of parameters. The results show a remarkably consistent pattern in that the components of freight demand that could change faster (i.e., freight production and attraction) are those that exhibit the largest rates of parameter change. The rates of change for these models are 18.29% and 26.37%, respectively. In contrast, the freight distribution models of loaded trips were found to change less rapidly (10.50% and 1.94%, respectively, depending on the impedance function), while the tonnage distribution model exhibited only fixed time effects. The model that changes least rapidly is the empty trip model, which has a rate of change of 0.83%.