TRANSIT – A model for simulating infrastructure and policy interventions in agriculture logistics: Application to the northern Australia beef industry
Introduction
Agriculture and horticulture in northern Australia is characterised by long supply chains between production and processing and markets in Australia, with transport distances often upwards of 2500 km. All year round access in the north is not possible due to a sparse road network that is regularly inaccessible in the wet season due to flooding or wet road conditions. The industry in the north is almost exclusively reliant on road for both business inputs and outputs. In the case of livestock, the northern beef herd of 12.5 million, across Queensland, Northern Territory and northern Western Australia (data based on Australian Bureau of Statistics Agricultural Commodities for 2012–2013, www.abs.gov.au) supplies nearly 90% of Australia’s live export cattle, with 694,000 head exported in 2011 (MLA, 2014). Non live export cattle are usually transported to feedlots and abattoirs in South Australia, Western Australia and Queensland. For example, nearly 50 per cent of cattle in the Northern Territory travel upwards of 1000 km between breeding property and abattoir (or port), with the transport cost exceeding A$150/head. Fig. 1 illustrates the location of northern Australia livestock enterprises in relation to the road network. A review of the northern Australian beef industry in terms of productivity and profitability (McCosker et al., 2010) indicated that a significant increase in costs of production has meant many properties are marginal and struggle during poor seasons. While the declining financial performance is largely a result of reduced real beef prices (Gleeson et al., 2012) as well as reduced turnoff and increased farm debt, McCosker et al. (2010) also identified rising overhead and direct costs such as freight as major contributors. The extensive spread of properties and declining financial performance is further complicated by market dynamics. Investment to support the resilience of the northern beef industry must anticipate and address future challenges, opportunities, and conditions. Infrastructure investments in roads, bridges and storage have the potential to substantially improve viability and resilience of the northern industry.
There is also an extensive range of cropping on irrigated land in the north (e.g. Ord River, Mataranka) and the high rainfall east coast. These include melons, mango, sugar, bananas, chickpeas and citrus. Produce is mostly trucked to markets and processors in southern Australia, at a distance of often greater than 2500 km and a cost of A$280/tonne. Whilst ports at Darwin, Wyndham and Townsville are a much shorter distance for exports, they provide limited facilities and economies of scale for refrigerated exports and storage. There is an anticipated major expansion of irrigated production in the north to over 110,000 hectares, and several cropping scenarios have been considered. To enable export supply chains through northern ports, infrastructure investments in the form of storage, processing and upgraded roads will be needed.
In this paper we outline a GIS based tool, TRAnsport Network Strategic Investment Tool (TRANSIT) that was designed to estimate the transport cost implications of investments and policies to enterprises across the agriculture supply chains from farm gate to processing or export. Road and rail transport are considered, including intermodal transfers, different vehicle combinations as well as regulatory constraints related to drivers and biosecurity.
The literature has a large range of methodologies to analyse and optimize transport logistics in agriculture supply chains. These include static analyses that map out existing chains to understand the performance of different segments of the supply chain, as well as to identify opportunities for increased efficiency and international competitiveness (Francis et al., 2008, Economic Associates, 2011, Uddin et al., 2011). The Council for Scientific and Industrial Research, CSIR (2005) conducted a major state of logistics study in South Africa aimed at defining infrastructure investment priorities across the major industry sectors including agriculture. In light of the high level analysis and recommendations from the work, several sector specific logistics projects between CSIR and South African agriculture industries were established. This includes a more detailed analysis on South African fruit logistics (van Dyk and Maspero, 2004) with a focus on providing recommendations for priority investments in infrastructure. The main disadvantage of these studies that analyse existing supply chains is that they do not allow alternative scenarios to be evaluated and compared.
Computer based models based on simulation and optimization have been used extensively in most agriculture industries to improve transport efficiencies for operational and seasonal planning. See Higgins et al. (2010) as well as Ahumada and Villalobos (2009) for reviews of such methods and applications. For strategic planning, the most common methods have been in location science. Location models have many potential applications for developing and optimizing scenarios for logistics infrastructure investment in agriculture, particularly to minimise transport costs. Lucas and Chhajed (2004) surveyed several past applications, which included location of sub-terminals for grains; processing plants for soybeans; rail-based grain distribution systems; feed and slaughter plants; and citrus packing houses. These older location model applications do not accommodate the complex geographical landscape of agriculture transport (e.g. road conditions, vehicle access) or the broader range of variables (energy supply, labour, biosecurity, access to amenities) related to the decision of where to locate an agriculture facility. In more recent research related to livestock strategic planning, Domingues-Zucchi et al. (2011) applied dynamic programming to optimize the location of abattoirs in Brazil accommodating transport costs and subject to a large range of spatial constraints.
The TRANSIT approach to modelling agriculture logistics scales up from of every vehicle movement between individual enterprises. This provides the capability to analyse a wide range of infrastructure or policy opportunities, whether small or large scale. There have been limited such ground-up attempts at modelling agriculture logistics. In cotton, Ravula et al. (2008) implemented discrete simulation to represent truck movements of cotton and biomass, to test management practices to increase utilization. A State of Logistics study by Higgins et al. (2011) aimed to develop and test a methodology that estimates the costs of logistics in Australian food industries and identify its drivers and challenges. Marquez et al. (2010) developed a freight flow model for all fruit and vegetable movements within as well as in and out of Victoria, with the goal of evaluating the costs of transport logistics under various scenarios of extreme weather events and price shocks. The broader agriculture logistics studies by Higgins et al. (2011) and Marquez et al. (2010) were restricted to small case studies or required heavy data interpolation on resource movements and routes, since detailed data sets on movements were not publicly available. A major advantage of the research to develop TRANSIT has been the ability to obtain and link the range of industry data sets to geographically map out individual supply chain movements across the industry.
TRANSIT was initially developed for the livestock industry in northern Australia, and was co-funded by the Australian federal and state governments (Queensland, Western Australia, Northern Territory) representing the north. Two workshops were held with government, industry and transport representatives to formulate the capability requirements of the model, which then helped define the methodology. The workshops revealed a broad range of capability requirements in identifying financial benefits to agriculture enterprises arising from:
- –
Upgrades of roads and bridges to accommodate larger combination vehicles and greater wet season accessibility.
- –
New road connections to reduce detours.
- –
Impacts of new facilities (processors, storage, cattle yards, driver stops) at different locations.
- –
Better utilization of rail transport and suitable hubs and loading points.
- –
Changes in bio-security regulations such as reducing some requirements of tick clearing. Livestock transported into a tick infested region must be treated at a facility to ensure they are not carrying any live ticks.
- –
Changes to driver fatigue regulations.
- –
Impact of extreme weather events on accessibility and transport costs.
Some of these applications (e.g. new abattoir or major highway upgrade) were very large scale and would benefit a large number of enterprises across more than one state, whilst other applications (e.g. minor bridge upgrade) would only benefit a small number of vehicle trips for a few enterprises. Through these applications, TRANSIT informs government and industry on infrastructure investment (and policy changes) by providing transparency on the transport benefits to each agriculture enterprise. The hypothesis behind TRANSIT was that industry wide efficiencies may be gained through a range of small changes and/or improvements to the network through strategic investment at critical locations. Financial benefits from more efficient transport may be accompanied by reductions in deterioration to the quality of the agriculture upon arrival at the destination, increased driver safety, and reduced emissions. There is the challenge that some enterprises will benefit a lot more than others and that agriculture transport often represents a small portion of some infrastructure usage (e.g. highways). A further challenge is “who pays” for infrastructure, whether it is a public or private investment.
Section snippets
Transit inputs
To develop and implement TRANSIT for livestock, data was gathered from more than 20 government departments and industry groups. Data were used for the TRANSIT model in three ways: constructing the routable transport networks, defining the vehicles and their travel features for the network; and constructing the livestock supply chains between different enterprises connected to the transport network.
Transit methodology
The goal of the TRANSIT engine is to optimize the transport route along the road/rail network for each O–D trip, and then calculate the cumulative impacts at the enterprise or regional scale by aggregating across the vehicle trips on each route. The current routes selected by vehicle drivers for transporting agriculture commodities between each origin and destination may not be the most efficient. Depending on a range of constraints at the origin, or along the transport route (e.g. flooded
Applications
Since the development of TRANSIT, it has been applied to several infrastructure and policy scenarios in northern Australia. These scenarios were identified by government and industry stakeholder groups and represent current major opportunities being investigated in the north. We demonstrate TRANSIT using three such case studies in this paper, and highlight how the analysis has been used by government and Councils.
Discussion and conclusions
Agriculture supply chains in Australia, particularly the north, are often characterised by long travel distances between production, processing and markets. This not only substantially increases the total cost of production, but also makes the industry vulnerable to market changes and climate events. It is well known in the industry that infrastructure investments and policy changes would address the transport challenges in agricultural supply chains. However, there was no tool available that
Acknowledgments
This work was co-funded by: Commonwealth Government – Office of Northern Australia; State Governments of Queensland, Western Australia and Northern Territory, Meat and Livestock Association, and Flinders Shire Council. A large amount of industry and state government support was provided to the project team in the form of local expertise, model requirements and key data sets. The authors thank the National Livestock Identification System for granting access to the historical movement data of
References (20)
- et al.
Application of planning models in the agri-food supply chain: a review
Eur. J. Oper. Res.
(2009) - et al.
Reprint of “Quantifying yield gaps in rainfed cropping systems: A case study of wheat in Australia”
Field Crops Res.
(2013) - et al.
Development of a system to produce maps of agricultural profit on a continental scale: an example for Australia
Agric. Syst.
(2012) - et al.
Cotton logistics as a model for a biomass transportation system
Biomass Bioenergy
(2008) Second annual state of logistics survey for South Africa: defining research priorities for developmental logistics
CSIR Build Environ.
(2005)National Greenhouse Accounts (NGA) Factors June 2009
(2009)A note on two problems in connexion with graphs
Numer. Math.
(1959)- et al.
Optimum location for export-oriented slaughterhouses in Mato Grosso, Brazil: a dynamic mathematical model
Int. J. Logis.: Res. Appl.
(2011) - Economic Associates, 2011. Development and assessment of options for the efficient and commercially sustainable...
- et al.
Value chain analysis in the UK beef foodservice sector
Supply Chain Manage.: Int. J.
(2008)
Cited by (11)
Life cycle perspective of bio-oil and biochar production from hardwood biomass; what is the optimum mix and what to do with it?
2019, Journal of Cleaner ProductionDynamic freight flow modelling for risk evaluation in food supply
2019, Transportation Research Part E: Logistics and Transportation ReviewCitation Excerpt :Burgholzer et al. (2012) also analyse detour traffics, but they use an event-driven, agent-based, intermodal traffic simulation to evaluate the effects of a disruption. Focussing on food products, Higgins et al. (2015) examine the vulnerability of the transport system of the Australian beef industry, which is characterised by particularly long transport distances. To this end, they develop TRANSIT (Transport Network Strategic Investment Tool), which calculates the optimal transport costs for different scenarios based on characteristics of the road network, regulations, vehicle types, and origin-destination pairs using functions of a geo-information system.
Informing transport infrastructure investments using TraNSIT: A case study for Australian agriculture and forestry
2018, Computers and Electronics in AgricultureCitation Excerpt :The Australian Local Government Association (ALGA) recommended a strategy that considers an ‘end to end’ supply chain approach (ALGA 2016). To respond to these national challenges, CSIRO developed the Transport Network Strategic Investment Tool (TraNSIT) (Higgins et al., 2015). This Tool provides a comprehensive view of transport logistics costs and benefits due to infrastructure investments and policy changes in agriculture and forestry supply chains in Australia.
Irrigated agricultural development in northern Australia: Value-chain challenges and opportunities
2017, Agricultural SystemsCitation Excerpt :Depending on the crop, the supply chain paths will vary in terms of segments (transport, processing, storage) and destinations, as illustrated schematically in Fig. 2. The Transport Network Strategic Investment Tool (TraNSIT) (Higgins et al. 2015) was used to create the supply chain for each regional case study and crop, to determine the logistics costs per tonne of product handled or for the whole case study region. While, TraNSIT is largely focused on transport costs, the other major supply chain cost elements of processing, packaging and storage were included where data were available.
Study on the Linkage Development of Logistics Industry and Agriculture: A Case Study of Kaifeng
2022, Proceedings - 2022 8th International Conference on Information Management, ICIM 2022