Informing macrologistics connectivity in emerging economies through a triangulated research approach:

the case of Uzbekistan

Written by: Dr Zane Simpson, Prof Jan Havenga, February 2022



Uzbekistan is Central Asia's most populous country, with its 32 million citizens representing nearly half the region's total population. Tashkent, the capital, is Central Asia’s biggest city. The government is exploring various initiatives for enhancing regional integration and connectivity, diversifying trade routes in the context of China’s Belt and Road Initiative (BRI), and exploring maritime access options. The country has the potential to become a regional hub linking Southeast Asia, South Asia, the Commonwealth of Independent States, and Europe (World Bank, 2020), where logistics and especially logistics facilities can play a key role in the region, similar to Singapore’s successful positioning as a hub for Southeast Asia (Yue & Lim, 2003). Uzbekistan’s economic development goal is to move from a centrally planned, inward-oriented economy towards a more open, integrated, value-adding and export-driven economy (Tsereteli, 2018) supported through reforms to further improve the investment climate, the efficiency of public sector investments, and service delivery mechanisms in various sectors of the economy.

The country’s ambitious New Development Strategy was launched in 2017 (Tashkent Times 2017) to support its transition to democracy, with the following key focus areas: state and judicial reform, economic liberalisation and growth, and social and cultural development and cohesion, including creating a foreign policy enabling regional cooperation. Economic liberalisation and growth include inter alia initiatives focused on the creation of industrial zones and technoparks, export facilitation and transport infrastructure development. These initiatives become even more pertinent given Uzbekistan’s double-landlocked status (all neighbours are also landlocked), making the country reliant on both infrastructure and relationships with neighbouring countries to reach international markets (Qoraboyev 2018). It is therefore imperative for Uzbekistan to develop a clear understanding of its own national logistics needs and priorities to leverage opportunities for connectivity created by the opening of the country and initiatives in the Central Asian region.

However, within Uzbekistan, the focus has been on the physical planning of transport infrastructure assets without explicitly considering the demand for freight transport and intermodal competition (World Bank, 2020). Accurate, timely, and reliable information is the foundation of sound national and regional logistics sector policymaking and investment decisions (Asian Development Bank, 2007); the absence of which remains a critical barrier to the sector's sustainability and environmental performance, especially in emerging economies (Moon, 2013).

One tool to step out of this negative cycle is to have the ability to target the sectors and logistics facilities which would not only enable short-term growth opportunities but would align with the economic development agendas of government and business over the long term. This is enabled through integrated data on the national transport and logistics industry with sufficient commodity and geographical disaggregation. Given the typical paucity of such data in emerging economies, this research describes the development of a freight-flow model for Uzbekistan – the Uzbekistan Freight-Flow Model (UFFM) – leveraging existing data sources. Both domestic flows within Uzbekistan, as well as freight-related movements beyond the borders of Uzbekistan, are included.

The UFFM was developed with support from the Government of Uzbekistan based on three drivers, i.e., the conviction that Special Economic Zones (SEZs) were required (Kuzieva, 2019); that the railway must improve (despite its sizable network, the railway only shipped 68.4 million tonnes in 2018, albeit up from 61.5 million tonnes in 2012 (Sultanovich, 2019)); and that a data-driven transport strategy for Uzbekistan is required. However, from exploratory discussions, given the country’s centrally planned economic history, it was clear that there would be challenges to readily obtain the detailed data required for the UFFM (further discussed in the methodology section). The primary aim was therefore to develop a freight-flow model with sufficiently detailed outputs to develop a cogent macrologistics narrative with evident application in the country’s development strategies. The resulting secondary aim was to pique the interest of stakeholders to obtain wider access to data sources to refine the UFFM.


Research approaches in data-scare environments

The United States Department of Transportation (2019) provides guidelines on data collection, compilation and refinement to support freight planning and forecasting (1) identify and assess available data sources and identify gaps; (2) collect, refine and clean existing data to create a base dataset of existing freight data; (3) collaborate with freight stakeholders to collect or estimate identified gaps; (4) process and combine existing and new data to create an integrated freight database; (5) create a process to maintain and update the freight database.

However, in emerging economies projects to improve data collection, accuracy and availability often fail because their goals are too ambitious, as the resources simply are not available to implement large-scale projects. This remains a key obstacle to the development of coherent national and regional transport policies and subsequent investments (Asian Development Bank, 2007).

To overcome the research limitations imposed by freight-flow data paucity, researchers often utilise the principle of data triangulation, i.e., the use of data from different sources to overcome the challenge of incomplete or conflicting datasets, to deepen understanding of the sector, and to cross-validate findings to increase the credibility of outputs (Mangan, Lalwani & Gardner 2004; Islam, 2005; Rahman, Mohammad, Rahim, Hassan, Ahmad & Kadir, 2017). The concept of triangulation refers to the trigonometric approach utilised in land surveying to determine unknown distances by measuring the angles in a triangle formed by three survey control points, i.e., two points on the baseline with known distance and a distant third point in line-of-sight (Shafer, 1987). Through various iterations utilising the newly developed data a chain of triangles or triangulation network can be developed to describe the landscape under survey. Outputs of the triangulation network are strengthened by increasing the observations or inputs (Moose & Henriksen, 1976). The important principles are that known data points can be utilised to estimate unknown data points, that estimates can be improved by utilising more data inputs, and that this iterative process can eventually lead to a reliable description of the whole landscape under investigation if done diligently. A further approach to enhance triangulation is the Pareto principle in that significant commodities and regions were investigated in more detail.

In a model to estimate maritime carbon emissions from international trade, Schim van der Loeff, Godar and Prakash (2018) address the historical lack of data on this subject by linking and integrating many data sources, previously used in isolation. Even in the case of developed economies, Müller, Wolfermann and Huber (2012) refer to the ‘scarcity of representative data’ to build a large-scale freight-flow model for Germany, ‘therefore, it is necessary to expand and specify the benefit of given data by skilful handling and combination’ (although, naturally, much more data is available than in emerging economies). Islam (2005) combined research methods (a literature survey, a quantitative survey, and a two-round qualitative Delphi study) to analyse the extent to which the transformation of a fragmented freight transport system into an integrated multimodal transport system depends on the present state of the country. The integration of several data sources is therefore a relatively common practice in freight-flow modelling at various levels of disaggregation due to the typical lack of data to address pressing macrologistics issues. In addition to the creative usage of available data, Chaberek and Mańkowski (2019) emphasise the need for “the right methods and tools” to develop a holistic map or model of the national logistics system for effective management of the sector. Schim van der Loeff et al. (2018) emphasise the value of spatially-explicit modelling to understand the causality between demand for a commodity from a specific origin and its associated logistics need and impact, both direct and with regard to externalities. This increases the policy relevance of freight logistics modelling.

The literature, therefore, informs two principles for application in the freight-flow modelling approach for Uzbekistan. Firstly, the use of a combination of input data sources that is verified through an iterative process of triangulation and, secondly, the need for spatially and commodity explicit modelling inputs and outputs to increase the macrologistics relevance of the modelling outputs. A hybrid or triangulated research approach is therefore adopted for the UFFM, with a specific emphasis on developing spatial and commodity characteristics of freight flows. This approach is detailed in the next section.



Uzbekistan freight flow model

The methodology for developing a freight flow model and related logistics costs model for Uzbekistan is a gravity model based on the supply of and demand for commodities within the economy (World Bank, 2020). The supply comprises local production and imports; while demand incorporates intermediate demand, final consumption, and exports. Due to the vast differences in the limited data available, the analysis aligned all the data to the major commodities in the customs data of UN Comtrade. 36 commodity groupings were then selected. Total supply and demand per commodity were then developed for each district. The total supply per district is depicted in Figure 1 and was developed for all commodities.


Figure 1 Uzbekistan

Figure 1. Total supply per district in Uzbekistan (left) and freight flows (right)


To obtain supply and demand data many different data sources from various local government and international agencies were triangulated. These included State Statistics data, road and rail data, customs and trade data, interviews and other sources.


Data limitations

As mentioned in the introduction, there were challenges to readily obtain the detailed data required for the UFFM. Firstly, not only is there a lack of data, but the legacies of the authoritarian structure of government remain a hindrance despite reforms (Omelicheva, 2016; Bowyer, 2018); data is often protected, and foreign involvement is questioned (Spechler, 2007). Secondly, Uzbekistan, being one of the only two double-landlocked countries in the world, relies heavily on its central position in Central Asia, but at the same time, many production factors are managed from outside the country. Landlocked countries normally face challenges to access world markets and lag behind neighbours in development and trade (Faye, McArthur, Sachs & Snow, 2004). For a double-landlocked country, where neighbours have similar problems, this situation is exacerbated. These factors further hinder the ability of researchers to gain access to relevant data.


Key descriptors of the Uzbekistan macrologistics landscape

Logistics Performance Index

The Logistics Performance Index (LPI) of a country is often quoted in transport industry reports and strategy documents as it is an acknowledged global dataset with improved methodology over time (Puertas & Garcia, 2014), and often in emerging economies, it is one of the few comparative statistics available. It is however a stand-alone assessment, often without contextual links and importance weighting of its components (Rezaei, Van Roekel & Tavasszy, 2018). Limitations of the measure include, specifically, skewed measurements if a country is a ‘victim’ of outside control of its logistics system, exacerbated by being landlocked (Beysenbaev & Dus, 2020). The LPI alone cannot inform a macrologistics strategy, but if used as a springboard for research to follow can play a useful role.

In 2018, Uzbekistan’s LPI position improved from number 118 to 99. Uzbekistan’s overall score of 2.58 is on par with the lower-middle-income country’s performance, with the country’s score increasing on all but one LPI indicator between 2016 and 2018. Timeliness and consignment tracking improved markedly which can be traced back to improved regulation. The drop in the score of efficient customs and border management clearance (‘customs’ in Figure 1) is significant, given the country’s dependence on freight outside of its borders (refer to Section ‎4.2). In 2018, the government took steps to reform these areas, including customs, and to open several border posts. The changes and impacts of these reforms might be captured in the 2020 LPI results. This problem is reflected in the reality of Uzbekistan’s freight flows which put a sharp focus on border and trade issues in a double-landlocked country, as discussed below.


Total freight flows

Based on the UFFM, freight demand for Uzbekistan within the borders of the country is 194.6 million tonnes and 77.6 billion tonne-kilometres (refer to Figure 1). This includes all domestic freight flows and flows towards border posts for exports, as well as from border posts for imports. Transverse freight is excluded.

If the freight flows to and from final destinations in foreign countries are added, freight demand increases from 60 billion tonne-kilometres to 138 billion tonne-kilometres (due to the added distance) (refer to Figure 3). Simply put, to achieve Uzbekistan’s final economic output 138 billion tonne-kilometres are required, 43% of which occurs outside of the country. This is exacerbated by the fact that cross-border land operations don’t have the very high-efficiency possibilities as maritime trade, where a handful of global maritime merchant companies have developed inordinate coordination skills over the last few decades. Added to this problem is Uzbekistan’s logistics service provider industry which is in its infancy and more or less all cross-border operations are by non-Uzbekistan companies. This effect is confirmed by Grafe, Raiser and Sakatsume (2008) who found that although regional market integration in Central Asia is quite high, borders do have a significant effect on price dispersion and that this effect is at its worst for Uzbekistan. Uzbekistan, therefore, has the highest border ‘friction’ in Central Asia, but at the same time is more dependent on cross-border land trade than all countries in Central Asia.

The second problem in Uzbekistan is the natural shape of the country which essentially splits the country into two parts: an Eastern portion east of Samarkand and a Western portion west of Samarkand. Freight-flow distribution is heavily skewed towards the East. The east has 65% of freight supply and 57% of freight demand, but only 13.5% of the land area (data from UFFM) (refer to Figure 2). At the same time, the Gross Regional Product of regions to the West range from 23 to 36 thousand Soums per capita whereas the figure for the East ranges from 27 to 93 thousand Soums per capita in 1995 comparable prices (Qayumovna, 2021).


Figure 2 Uzbekistan

Figure 2. Uzbekistan freight outside of the border (left) and East-West split (right)


From these observations, clear overarching macrologistics objectives for Uzbekistan emerge. A logistics action plan should consider development objectives in the Western part of the country and strategies to streamline border crossings and exposure to freight risks beyond the borders due to the inordinate long distances of cross-border flows. This is therefore a spatial development problem and the absence of a powerful logistics hub to control logistics not only in the country but in the region.


Concept analysis – domestic opportunities for improving

Two key options were investigated, i.e., improved use of the railway, and the clustering of freight through the development of SEZs and, ultimately, freight villages. (1) The improved use of the railways can reduce transport costs, improve Uzbekistan’s positioning in Central Asia and make it easier for the West to access markets; and (2) logistics centres (hubs or SEZs) can also through consolidation improve the locus of control of Uzbekistan to leverage the BRI. The high-level outputs of these options are discussed below.


Improved use of the railway

Uzbekistan is relatively ‘oversupplied’ by rail if comparing the geographical and economic output size of the country with the rest of the world. Three avenues for improved use of the railway are considered namely modal shift, savings due to increased rail density, and savings due to improved rail efficiencies. Using the UFFM it is estimated that around 21.9 million tonnes of freight or 17.9 billion tonne-kilometres can shift to rail, resulting in a transport cost saving of US$ 430 million (due to lower rail costs). This shift will densify Uzbekistan’s rail network which would result in a further decrease in the costs of Uzbekistan rail freight by 17%.  

Although not enough data is available for an accurate measurement, observations around margins and efficiency indicate that US$ 180 million could be saved by making the railway more efficient, cost-effective and with more sustainable and fair margins.


Clustering freight

Clustering freight in SEZs and, ultimately, freight villages, will have an important effect on Uzbekistan’s logistics costs. Freight villages shorten distances in the supply chain, enable more accurate delivery windows, and consolidate long-distance freight which facilitates modal shift and decreases the unit cost of both road and rail transport. Investment in a refined freight-flow model will enable improved freight village positioning and design.

It seems as if only 5% of current freight flows will be affected by the current design (of logistics centres or logistics hubs), but an improved design where between 10% and 20% of freight could be captured, is potentially achievable and would improve the success of freight villages.


Dependence on trade routes and logistics outside of the country

Most important, however, is Uzbekistan’s exposure to the regional freight network design. If tonne-kilometres of Uzbekistan freight outside of the country is added to domestic freight the volume grows by 60 billion tonne-kilometres from 78 to 138 billion tonne-kilometres. As a comparison, South Africa’s 350 billion tonne-kilometres constitute only 20 billion tonne-kilometres added by overland or surface cross-border freight. Countries such as South Africa require much more maritime freight for trade, but this freight is already, for the most part, efficiently consolidated, globally scheduled and part of an efficient global shipping line system. These conditions do not exist for a double-landlocked country and need to be created by a regional transport system that is run efficiently, similar to a global port and shipping line network. In terms of transport costs, also, this characteristic of the Uzbekistan freight system means that more money (52% of transport costs) is spent on freight transport costs outside of the country than within the country.


Recommendations and action plan

Olofin, Olubusoye and Salisu (2011) emphasise that an accurate, timely and systematic national statistical system is a prerequisite for national development (i.e., the fulfilment of a country’s objectives), where this ‘system’ incorporates the people, procedures, data, and equipment to bring it to fruition. As one of the backbones of a globalised economy, the expansion of national statistical systems to include freight logistics intelligence, is long overdue. The research highlighted key principles to support both the development and application of such national statistical systems are: a culture of research-driven policy decisions; establishment of an active data users’ forum; development of robust models; capacity building; and enactment of appropriate legal frameworks.

The following recommendations and action plan for Uzbekistan support these principles and are informed by the outputs of the UFFM:

  1. Manage transport and logistics as a strategic commodity. This means that the Transport Ministry should, on the one hand, be capacitated adequately and given enough powers to effectively develop strategic solutions while, on the other hand, integrative policy development with the ministry involved should receive attention.
  2. Firm up on data and statistics. The poor state of statistics has many aspects. Data is not correctly captured, collated, and published, discrepancies are not questioned or cannot be explained, and data is often quoted without interpretation or context. A strategy is required to develop a set of reliable and useful statistics that allows for national policy development and decision-making. An activity-based freight flow model for the country, building on the UFFM, will go a long way towards solving this problem. Importantly, this model should have a 30-year forecast component.
  3. Data-driven joint strategy development. Policymakers should involve all stakeholders in development discussions, based on analysed statistics, to inform decision-making. There are many suggestions on how to improve the freight-flow landscape, but a rough analysis points to different priorities for improvement. As an example, streamlining border crossings and the development of a domestic logistics industry that can work across borders is often mentioned, but improving clustering and developing a regional intermodal strategy might be more important.
  4. Consolidation centres that are local- and region-critical. Little evidence could be found that the positioning and design of the current SEZs will be optimal. This can only be determined by a freight flow and related cost model. Evidence from other countries point towards this being one of the most important spatial and logistics strategies and should be informed by more in-depth freight-flow analysis and forecasts.
  5. Insert rail effectively into domestic and regional supply chains. This relates to linkages with SEZs, the design of the SEZs, rail’s role in regional intermodal and the efficiency of the railway itself.
  6. Become more effectively involved in a regional freight flow, clustering, and intermodal strategy. Leadership in evidence-gathering (both from other nations and through proper development of its own statistics) will allow Uzbekistan to take a lead in a Central Asia freight strategy. Maritime nations have the advantage of a highly efficient, global water-based trade system. Uzbekistan’s inordinate reliance on surface freight is a big risk.
  7. Develop a logistics strategy that specifically concentrates on the Western portion of the country. Development ideals for the West must be confirmed and strengthened. This should be forecasted in terms of freight flows to align logistics strategies with freight flows.


Concluding remarks

The concept UFFM is the starting point of a macrologistics decision-making tool and can be refined with access to outstanding data sources. Further refinement is advised to aid more detailed industry and regional analysis and scenario development for the identification of priorities to address modal, spatial and regional integration challenges which, in turn, will improve logistics costs and support the shift from fragmented to integrated multimodal planning.



Asian Development Bank. 2007. Oceanic Voyages: Shipping in the Pacific, Pacific Studies Series. Manila: Asian Development Bank.

Bowyer, A.C. 2018. Political Reform in Mirziyoyev’s Uzbekistan: Elections, Political Parties and Civil Society. [Stockholm, Washington, DC]: Silk Road Studies Program, Central Asia-Caucasus Institute.

Chaberek, M. & Mańkowski, C. 2019. A methodological approach to modelling national input-output logistics flows. Prace Naukowe Uniwersytetu Ekonomicznego we Wrocławiu, 63(12):179-195.

Faye, M.L., McArthur, W.M., Sachs, J.D. & Snow, T. 2004. The Challenges Facing Landlocked Developing Countries. Journal of Human Development, 5(1):31-68.

Grafe, C., Raiser, M. & Sakatsume, T. 2008. Beyond borders—Reconsidering regional trade in Central Asia. Journal of Comparative Economics, 36(3):453-466.

Islam, D.M.Z. 2005. International freight transport multimodal development in developing countries: the case of Bangladesh. Published dissertation. Plymouth: University of Plymouth.

Kuzieva, N.R. 2019. Free economic zones of Uzbekistan: development, operation and features of the tax regime. International Journal of Research in Social Sciences, 9(4):1173-1183. 

Mangan, J., Lalwani, C. & Gardner, B. 2004. Combining quantitative and qualitative methodologies in logistics research. International Journal of Physical Distribution & Logistics Management, 34(7):565-578.

Moon, J.R. 2013. Strengthening Inter-island Shipping in Pacific Island Countries and Territories. [Geneva]: United Nations ESCAP, SPC CPS, PIFS, IMO.

Moose, R.E. & Henriksen, S.W. 1976. Effect of Geoceiver Observation Upon the Classical Triangulation Network. [Maryland]: NGS, NOS

Müller, S., Wolfermann, A. & Huber, S. 2012. A nation-wide macroscopic freight traffic model. Procedia - Social and Behavioural Sciences, 54(4):221-230.

Olofin, S.O., Olubusoye, O.E. & Salisu, A.A. 2011. The place of data and models in the planning and transformation of the Nigerian economy. The Nigerian Journal of Economic and Social Studies, 53(3):219-246.

Omelicheva, M.Y. 2016. Authoritarian legitimation: assessing discourses of legitimacy in Kazakhstan and Uzbekistan. Central Asian Survey, 35(4):481-500.

Puertas, M.L.R. & García, L. 2014. The importance of the Logistics Performance Index in international trade. Applied Economics, 46(24):2982-2992.

Qoraboyev, I. 2018. The Belt & Road Initiative and Uzbekistan's New Strategy of Development: Sustainability of Mutual Relevance and Positive Dynamics. Uzbek Journal of Legal Studies, 2:30-43.

Qayumovna, J.Z. 2021. Modern trends in the economic development of the regions of Uzbekistan, Scientific horizon in the context of social crises. Scientific Collection Interconf, 41:56-62.

Rahman, N.A.A., Mohammad, M.F., Rahim, S.A., Hassan, R., Ahmad, M.F. & Kadir, S.A. 2017. Shipper's perceptions of aviation Logistics Service Quality (LSQ) of air freight provider. Journal of Engineering and Applied Sciences, 12(3):699-704.

Rezaei, J., Van Roekel, W.S. & Tavasszy, L. 2018. Measuring the relative importance of the logistics performance index indicators using Best Worst Method. Transport Policy, 68:158-169.

Schim van der Loeff, W., Godar, J. & Prakash, V. 2018. A spatially explicit data-driven approach to calculating commodity-specific shipping emissions per vessel. Journal of Cleaner Production, 205:895-908.

Shafer, M.L. 1987. Triangulation, in R.C. Brinker & R. Minnick (eds.). The Surveying Handbook. Boston: Springer. 296-339.

Spechler, M.C. 2007. Authoritarian politics and economic reform in Uzbekistan: past, present and prospects. Central Asia Survey, 26(2):185-202.

State Statistics Committee of the Republic of Uzbekistan. 2019. Correspondence. 29 March, Stellenbosch.

Sultanovich, F.J. 2019. Improvement of Economic Efficiency of Development of Railway. Asian Journal of Technology & Management Research, 9(2):57-60.

Tashkent Times. 2017. Uzbekistan's Development Strategy for 2017-2021 has been adopted following public consultation [Online]. Available: [2021, February 15].

Tsereteli, M. 2018. The economic modernization of Uzbekistan, in S. Frederick & S.E. Cornell (eds.). Uzbekistan’s new face. Maryland: Rowman & Littlefield. 82-114.

UN Comtrade. 2019. United Nations Statistics Division [Online]. Available: [2019, June 1].

United States Department of Transportation. 2019. Quick response freight methods – September 2019. Washington, DC: United States Department of Transportation.

World Bank. 2020. Uzbekistan - Building Blocks for Integrated Transport and Logistics Development: Policy Paper. Washington, DC: World Bank Group.

Yue, C.S. & Lim, J.J. 2003. Singapore: A regional hub in ICT, in S. Masuyama & D. Vandenbrink (eds.). Towards a Knowledge-based Economy: East Asia's Changing Industrial Geography. Tokyo: Nomura Research Institute. 259-298.