Application of a multi-criteria decision-making technique to prioritize the irrigation projects in southern Nepal Open Access

Agriculture is the primary commodity of most developing countries, providing livelihoods and employment opportunities for people. In Nepal, about 66% of the population depends on agriculture, yet the sector contributes only 20% to the gross domestic production while employing 65% of the labor force (MoF 2021). Despite having 4.12 million hectares (ha) of cultivable land (MoALD 2022), only 39% of this area receives year-round irrigation (DoWRI 2019a), leading to agricultural instability and increasing reliance on imports. In 2022, agricultural imports surged by 18% from the previous year, overshadowing exports by a factor of seven (MoALD 2022). Several factors compound these challenges: high population density, limited cultivated area, traditional farming practices, rising pesticide costs, fertilizer shortages, and the impact of climate change all contribute to low agricultural productivity (Kumar et al. 2022; Shanmugavel et al. 2023). Furthermore, low irrigation efficiency, siltation in canals and fields, unmanaged irrigation practices, low investment in the irrigation sector, and lack of regular operation and management are the key challenges for the development of reliable irrigation facilities (Pradhan & Belbase 2018). Rapid urbanization has exacerbated the situation by encroaching on agricultural land and increasing food demand. The growing disparity between agricultural supply and demand jeopardizes both the national economy and food security (Kang et al. 2023). Addressing these issues requires expanding year-round stable irrigation and implementing robust systems supported by proper operation and regular maintenance.

The Terai region is the southernmost part of Nepal, consisting of fertile alluvial plains that support the majority of the country's agricultural activities. The region is characterized by a subtropical climate with high seasonal rainfall, predominantly during the monsoon season. The flat topography, combined with the presence of three major Himalayan river systems (Koshi, Gandaki, and Karnali), makes it a crucial agricultural hub. However, irrigation remains a major challenge due to sedimentation, inefficient water distribution, and inadequate infrastructure. Only a small fraction of it is used for irrigation in the Terai region. Among the available agricultural land in Terai, only 18% receives year-round irrigation (DoWRI 2019a). The complex terrain and substantial investment costs make it difficult to harness water from these major river systems leading to the reliance on medium- and small-sized rivers (Bista et al. 2021), groundwater, and erratic rainfall. Although 25 irrigation systems with a command area exceeding 25 ha are operational in the Terai, covering a net command area of 303,799 ha, the canal network is deteriorating, and hydraulic structures have exceeded their design life. Silt deposition in canals further disrupts water distribution and reduces operational efficiency. Because of gravity-fed irrigation systems, water is typically diverted into canals using overflow weirs or barrages, which regulate the flow from rivers. While these structures ensure a continuous water supply, they also introduce significant amounts of sediment into the canal network. Excessive sedimentation leads to canal clogging, reduced carrying capacity, and decreased irrigation efficiency. Regular sediment management through desilting and controlled flushing is essential to maintain canal performance. Sedimentation affects canal carrying capacity by reducing the cross-sectional area of flow. A heavily silted canal may operate at only 50% of its design discharge, limiting the area that can be irrigated. Desilting programs can restore flow capacity, potentially increasing the command area by up to 20% without additional infrastructure investment. This suggests that projects like Rajapur and Narayani, which suffer from high sedimentation, could benefit significantly from targeted desilting initiatives.

With limited financial resources, constructing new structures for efficient distribution, silt removal, and complete rehabilitation is prohibitively expensive. Consequently, selecting the most effective irrigation system for maximizing agricultural productivity has become increasingly challenging under these constraints. While constructing a new irrigation infrastructure is costly, regular maintenance and modernization are necessary for sustaining irrigation efficiency. Many international donor agencies prioritize investment in large-scale projects where the cost-benefit ratio is high. However, targeted interventions, such as selective infrastructure rehabilitation and improved sediment management, can significantly enhance irrigation efficiency without requiring prohibitively high capital investments.

The government of Nepal has an ambitious goal to increase agricultural production by 80% by the end of 2030 (DoWRI 2019b). To achieve this objective, the number of irrigation projects has to be increased, making the effective operation and maintenance (O&M) of the existing systems essential. Balancing new investments with O&M costs is crucial to ensure the sustainable development of irrigation infrastructure (Mwendera & Chilonda 2013). These O&M costs can be achieved either through government subsidies or water taxes from the farmers. However, with both mechanisms not functioning effectively, the budget for maintenance remains limited. On the other hand, poor maintenance reduces the active life of a project. Given the need to strengthen and expand the existing agricultural land, project prioritization becomes a critical element in investment planning and the sustainable development of irrigation facilities.

The overall performance of the project depends on various performance indicators like efficiencies, moisture content, soil type, and infiltration rate, among others. A project achieves higher performance and greater agricultural production if individual criteria are met effectively, justifying a higher priority for that project. The selection of these criteria is based on the evaluation metrics for investment and O&M, which are crucial factors in the decision-making process. In this context, MCDM techniques provide a powerful framework for evaluating various factors in decision-making (Köhler et al. 2019). These techniques enhance transparency, auditability, and analytic rigor in decision-making (Dunning et al. 2000). While several qualitative and quantitative methods can be employed to prioritize projects, MCDM techniques are particularly effective when reliable information is available. This approach evaluates multiple alternatives based on different criteria to select the best option. Consequently, MCDM techniques have been extensively used to develop strategies for agricultural water management (Radmehr et al. 2022), prioritize and select agricultural irrigation systems (Veisi et al. 2023), prioritize adaptation measures for agriculture (Acharjee et al. 2020), identify erosion-prone watersheds (Sarkar et al. 2022), manage sustainable energy (Arayeh 2015), delineate potential groundwater recharge areas (Lamichhane & Shakya 2019), and map flood hazards (Mudashiru et al. 2022).

These studies focus on prioritizing projects based on different alternatives and selecting the best option for planning and development. However, the application of MCDM techniques for prioritizing irrigation projects based on O&M remains limited. The objectives of this study are to (i) collect data on irrigation criteria of five projects through field survey, testing, measurement, and stakeholder interactions, (ii) conduct a questionnaire survey with expert consultations, and (iii) prioritize the best irrigation project based on criteria weights using the analytical hierarchy process (AHP). This method involves pairwise comparison of the alternatives for each criterion, followed by aggregation to compute the overall score (Saaty & Vargas 2022). This study addresses the dearth of MCDM-based project prioritization in the context of irrigation O&M, providing insights for more effective investment planning and irrigation development in Nepal.

RIP is a gravity-flow irrigation system located in Bardiya district, Nepal. It covers a command area of 14,880 ha and is situated between the Karnali River to the west and the Giruea River to the east, as shown in Figure 1(a). RIP is one of the largest farmer-managed irrigation systems in the region, relying primarily on the flow from the Giruea River flow. Most of the cross and head regulators of the project were found to be ungated, affecting its operational efficiency.

KNLIS with a command area of 250 ha is located in Banke district, as shown in Figure 1(b). Kiran Nala River is the primary source of irrigation for KNLIS.

NIS has the command area of 37, 400 ha, located in Parsa district, as shown in Figure 1(c). Narayani River is the primary source of irrigation for NIP, with a barrage constructed to divert water into the system. The eastern canal, 92 km long with a capacity of 850 m³/s, exhibits signs of deterioration, resulting in only approximately 30% of the command area receiving consistent year-round irrigation. Notably, heavy silt deposition near the head regulator and the absence of effective silt exclusion measures along the main canal exacerbated the issue. Although a silt ejector was installed at the head race of the main canal, its functionality was compromised, contributing to inadequate silt removal.

MISP1, located in Sarlahi district, has a command area of 2, 214 ha, Figure 1(d). The system depends on the Manushmara River, a spring water source, for irrigation. The canal network and hydraulic structures have deteriorated and passed the design life. Large amounts of silt were also observed along the bed and side of the canal, further impeding water flow.

KHIS is one of the largest public irrigation systems, with a command area of 25,000 ha in the Siraha district, Figure 1(e). It was primarily designed to provide supplemental irrigation for monsoon paddy. The Kamala River, a perennial river, is the primary source of irrigation for KHIS. Like the other systems, the canal and hydraulic structures deteriorate over time, with significant silt deposition observed along the canal bed and sides.

The performance of an irrigation project depends on several criteria, including conveyance efficiency, water application efficiency, soil type, infiltration rate, soil moisture, evapotranspiration, crop intensity, and command area. To comprehensively assess these factors, we conducted field measurements and tests as the primary step. Subsequently, a questionnaire survey was undertaken, engaging diverse groups of stakeholders, including farmers, officials, engineers, consultants, and other relevant experts. Their insights and expertise were collected to assign appropriate scores, on a scale of 1–9, to each parameter, thereby capturing the ground truth. The average score for each parameter was then utilized for further analysis (Table 3).

The command area data for the projects were collected from the Water Resource Research and Development Centre, Nepal.

Soil type influences the amount of water that can be retained in soil per unit depth. Sandy soil can store a significantly smaller amount of water for a short period compared to clay soil, demanding more frequent water application. The soil characteristics of the irrigation projects were assessed through field observation and surface exploration. Sieve analysis and soil gradation tests were conducted on the soil type in each project area.

Evapotranspiration is the loss of water due to evaporation from the soil surface and transpiration through the crop. It is influenced by weather conditions (solar radiation, air temperature, humidity, and wind speed), crop factors (crop type, ground coverage, and root density), soil characteristics (moisture and salinity), and management practices (mulching, shading, and fertilizer application). Accurate estimation of evapotranspiration is essential for planning, designing, and operating irrigation systems. The Penman–Monteith method (Monteith 1965) was used to estimate the potential evapotranspiration. It operates on solar radiation, air temperature, relative humidity, and wind speed. These data were collected from the Department of Hydrology and Meteorology, Nepal.

Cropping intensity refers to the number of planting cycles within a year (Pan et al. 2021). It is the primary factor affecting crop production and agricultural intensification (Biradar & Xiao 2011). Cropping intensity is calculated as the ratio of the gross command area to the net sown area. An increase in cropping intensity boosts crop production and enhances the farmer's economic returns.

Seventeen questions (Supplementary S1) were prepared and distributed among the experts, including engineers, project managers, farmers, and other stakeholders, to assign scores to the field criteria. The comparison judgment scale proposed by Saaty (1980) was used for the assignment of scores to the criteria. The scale ranges from 1 to 9, where 1 indicates equally important and 9 indicates the most important. The reciprocal value (1/1–1/9) indicates the least importance. The questionnaires and criteria are available in the Supplementary Section (S1).

MCDM is used to prioritize alternatives based on multiple criteria, making it particularly useful in decision-making where various parameters influence the outcome (Triantaphyllou 2000; Ozsahin et al. 2021). Since these parameters may have different units, normalization is applied to make values comparable. We employed the linear maximum and sum-based normalization methods, as suggested by Vafaei et al. (2016), ensuring that the sum of each column equals one (Saaty 1980). Logarithmic normalization was avoided due to its potential to produce zero or infinite values, making it unsuitable for the AHP method.

AHP, a widely used MCDM technique, structures decision-making into a hierarchy where weighted factors determine prioritization (Saaty 1980). The process involves (i) forming a pairwise comparison matrix based on expert scores (ranging from 1 to 9), (ii) normalizing data into unit-less values, (iii) assessing consistency to minimize bias, and (iv) ranking alternatives based on their relative importance (Saaty 1980; Saifullah 2019). The assigned weights significantly impact the final prioritization outcome.

The value of RI depends on the number of criteria selected for the study. For this study, we used nine criteria, corresponding to the RI value of 1.45 suggested by Saaty (1980). The value of CR less than or equal to 10% is considered as rational.

Nine criteria were chosen based on previous studies. These parameters were measured in the field for different irrigation projects. The difference between each criterion and irrigation performance and productivity is presented in Table 1. NIS is the largest irrigation project, characterized by dominant clay soil with the lowest infiltration rate (3.66 mm/h) and conveyance efficiency (41%), among the selected projects. It has a cropping intensity of 105% and a water application efficiency of 48%. KNLIS is the smallest project but demonstrates the highest conveyance efficiency (67.26%) and cropping intensity (147%). Clay loam soil with an infiltration rate of 5.57 mm/h is dominant in this project. RIP is dominated by sandy soil with a high infiltration rate (61.50 mm/h). Conveyance efficiency directly impacts the volume of water available for irrigation. For each additional percentage point increase in conveyance efficiency, there is a proportional increase in the volume of water delivered to fields, which can lead to a significant rise in crop yields. For instance, an increase in water availability from 40 to 60% efficiency could result in a 30–50% improvement in crop productivity, depending on soil and climate conditions. Also, sedimentation affects the canal carrying capacity by reducing the cross-sectional area of flow. A heavily silted canal may operate at only 50% of its design discharge, limiting the area that can be irrigated. Desilting programs can restore flow capacity, potentially increasing the command area by up to 20% without additional infrastructure investment. This suggests that projects like Rajapur and Narayani, which suffer from high sedimentation, could benefit significantly from the targeted desilting initiatives. The summary of the field measurement is shown in Table 2.

The average score by experts on nine field criteria during the questionnaire survey is presented in Table 3. It indicates the importance of criteria with respect to others. We conducted a questionnaire survey involving twenty participants (officials, engineers, consultants, farmers, and others) to acquire diversified information for generating the pairwise comparison matrix.

The pairwise comparison matrix was developed (Table 4), and reversal was done with the respective criteria for the application of the AHP method. For instance, conveyance efficiency is considered nine times more important than the command area. We then developed a normalization matrix to rank the alternatives, as shown in Table 5. We used CR to assess the suitability of normalization techniques and compared it with the table value using a linear normalization technique. The analysis resulted in a CI of 0.14 based on the comparison matrix and the normalization matrix. The CR is calculated as 0.09 (<0.10, within the acceptable limit). These results indicate that the pairwise compression matrix and normalization matrix provided consistent output among the criteria, and the generated weights of each criterion provided realistic information.

The weighted fraction of selected criteria is shown in Table 5. Efficient water utilization in the field is the main indicator of irrigation project performance and its impact on agricultural productivity. Consequently, conveyance and application efficiency are the key indicators for irrigation project selection. Studies show that about half of irrigation water could be lost through this process. Crop production increases exponentially with the increase in efficiency. Therefore, these criteria have high priority. Similarly, agricultural productivity depends on moisture content and infiltration capacity. Soil type, moisture content, and infiltration rate are strongly linked and control extra water losses in the field compared to efficiency. These criteria have a moderate weight for the irrigation project. Evapotranspiration is basically dependent on the meteorological characteristics of the command area that are out of human control. Increased evaporation reduces the project performance. Cropping intensity reflects farmers’ practices according to their needs, while the command area relates to land availability with respect to water availability. Based on expert judgment, these criteria have lower preferences compared to others. The result showed that conveyance efficiency (0.3) is identified as the most important criterion and ranked first, followed by water application efficiency (0.21) and soil type (0.15), respectively. The command area, with a weighted factor of 0.02 is identified as the least important criterion. These weights significantly influence the prioritization of the alternatives.

Irrigation projects are prioritized as shown in Table 6. Priority is given to the project that scores more than 60%. The Kiran Nala Lift Irrigation Project (70%) was found to have higher priority, mainly due to its high conveyance and application efficiency, low evapotranspiration rate, and appropriate soil characteristics of the system compared to others. Conversely, the RIP (51%) has less priority. Other projects have good overall scores and can be chosen for operation and maintenance.

Among several methods, we employed the commonly used AHP method for the MCDM technique. The details of other methods can be found in Vassoney et al. (2021). Each method has its advantages and limitations. The reliability of the selected method depends on several factors. A biased opinion of an individual or a team member may distort the judgments, affecting the results (Saaty & Vargas 2022). Additionally, the variability in judgements may result in inconsistent outcomes.

The AHP is a data-extensive method; collecting data can be challenging to implement in developing countries with limited resources. The weights assigned for each criterion are often subjective, can introduce uncertainty, and affect the outcomes. Generally, the AHP method is well suited for small and simple projects due to the low number of criteria and alternatives involved.

We presented a novel approach to prioritize the irrigation projects in Nepal using a MCDM technique. This study addresses a critical gap in irrigation management by focusing on O&M prioritization with limited financial resources.

We conducted a comprehensive assessment of five irrigation projects in the Terai region: RIP, KNLIS, NIS, MISP1, and KHIS. These projects were evaluated using nine key criteria: conveyance efficiency, water application efficiency, soil type, infiltration rate, initial and final soil moisture contents, evapotranspiration, cropping pattern, and command area. These criteria were carefully selected based on their significant impact on irrigation performance and agricultural productivity.

In the AHP analysis, the formation of the comparison matrix, normalization matrix, and the obtained CR value (less than 10%) indicate that the assigned weights for each criterion are consistent. The analysis revealed that conveyance efficiency had the highest weighted factor (0.30), highlighting its critical role in irrigation performance, while the command area had the lowest weighted factor (0.02), signifying its relatively lesser influence.

Among the evaluated projects, the Kiran Nala Lift Irrigation Project was identified as the highest priority for operation and maintenance, scoring 70%. Despite its smaller command area, its efficient water use and minimal losses make it a strong candidate for investment. In contrast, the RIP was ranked as the lowest priority, scoring 51%.

Furthermore, conveyance efficiency (0.30) and water application efficiency (0.21) emerged as the most critical factors influencing irrigation project performance. Projects suffering from high infiltration losses and siltation require targeted interventions for rehabilitation to maximize their benefits. Conversely, the command area was found to have the least influence in the prioritization process.

This study presents a robust framework for prioritizing the irrigation projects in Nepal using the AHP method, emphasizing conveyance and application efficiency as key determinants of irrigation performance. By prioritizing projects based on efficiency and infrastructure health, policymakers and irrigation managers can allocate resources more effectively, enhancing decision-making for sustainable irrigation development.

The proposed methodology is adaptable to other irrigation systems facing similar challenges, ensuring optimal resource utilization in water-scarce regions. Additionally, this framework supports the prioritization of both ongoing and prospective projects while aiding in the O&M of the existing systems. Although this study focused on five projects and seventeen questionnaires, its approach remains applicable to a broader range of irrigation projects. Future studies may integrate additional socio-economic factors to refine project prioritization further.

The authors are thankful to the Department of Hydrology (DHM) for providing the daily precipitation and temperature data. The authors are grateful to the Water Resource Research and Development Centre (WRRDC) for sharing the data of the command area for rogation projects.

The authors declare that they have no known competing financial interests or personal relationships that have appeared to influence the work reported in this paper.

All relevant data are available from an online repository or repositories.

The authors declare there is no conflict.