Production Systems Simulator Research Articles

Modeling long-term nitrogen utilization under alfalfa–corn rotation in Northeast China

ContextNortheast China is the main commodity grain base in China owing to its intensive corn (Zea mays L.) planting and heavy nitrogen (N) application. Alfalfa (Medicago sativa L.) and corn rotation have the potential to decrease N application and enable efficient N utilization, but its long-term impacts remain unclear. ObjectiveThis study utilized Agricultural Production Systems sIMulator (APSIM) to evaluate the N process of long-term alfalfa–corn rotation, including the soil N status and N legacy benefits of subsequent corn crops, and to optimize alfalfa–corn rotation. MethodsAPSIM was calibrated and validated utilizing datasets of crop N uptake (Nupt), soil inorganic N, and soil total nitrogen (TN) under continuous corn, continuous alfalfa, and rotations of three, four, and five years of alfalfa followed by two years of corn, defined as CC, CA, and 3A2C, 4A2C, and 5A2C treatments, respectively, which were collected from a field experiment in Quanyangou–Lishu, Jilin Province, China during 2014–2022. The validated APSIM was then utilized to simulate long-term (1981–2020) effects on N utilization under CC, 3A2C, 4A2C, and 5A2C treatments at Quanyangou–Lishu, Fujia–Lishu, Chaoyang, and Hailun in Northeast China. ResultsAPSIM effectively simulated the Nupt, inorganic N, and TN, with normalized root mean square errors of 5%–19%, 20%–30%, and 2.0%–4.2%, respectively, between the observed and simulated values. Compared with the CC treatment, alfalfa–corn rotation treatments reduced the N fertilizer application by 143 kg N ha−1 and significantly improved the net N mineralization and Nupt by 41 and 107 kg N ha−1, respectively, achieving increases of 50% and 517% in N use efficiency and N partial factor productivity based on corn equivalent yield, respectively. Alfalfa–corn rotation also consumed 94% of 200–500 cm inorganic N through alfalfa roots absorbing N after 9–14 years, which decreased the risk of soil inorganic N accumulation and N leaching. After 40 years, SON values of alfalfa–corn rotation treatments were 8% and 4% higher in the 0–20 and 20–60 cm soil layers and 1.6% lower in the 60–200 cm soil layer than those of the CC treatment at Quanyangou–Lishu, Fujia–Lishu and Hailun. The SON values of alfalfa–corn rotation treatments were 2.9% lower in three soil layers at 0–200 cm deep relative to those of the CC treatment at Chaoyang, with low precipitation of only 478 mm. Alfalfa–corn rotation also improved the first-year corn yield following alfalfa by 2.3%–8.8% at Quanyangou–Lishu and Fujia–Lishu. ConclusionsThe 4A2C treatment at Quanyangou–Lishu, Fujia–Lishu, and Hailun and the 3A2C treatment at Chaoyang are recommended in Northeast China to increase N use efficiency and N partial factor productivity by decreasing N fertilizer application, increasing SON, and absorbing inorganic N from the 200–500 cm soil layer during the alfalfa growing period and increasing mineral N during the corn growing period. ImplicationsOptimal alfalfa–corn rotations show great potential to reduce soil inorganic N residue and increase N use efficiency and SON in Northeast China.

A framework for assessing the impacts of land-use/cover change and climate change on wheat productivity under 1.5 and 2.0 °C warming at watershed scale.

Irrigation is used extensively to enhance grain production and ensure food security. Many studies have used crop models and global climate models to study the variation of irrigated crop yield in the context of climate change. But most considered the influence of direct climate change but neglected the influence of irrigation water availability, which is affected by land-use/cover change (LUCC) and indirect climate change, on irrigated crop yield. This study therefore developed a framework including Patch-generating Land Use Simulation model, Soil and Water Assessment Tool, Agricultural Production Systems sImulator Model, and global climate models for exploring the impacts of LUCC, direct climate change, and indirect climate change on wheat yield in a typical watershed. Both LUCC and climate change caused increased runoff from October to May, and thus increased the irrigation water availability, by 51.6 and 30.7 mm per growing season under 1.5 and 2.0 °C warming, respectively. The combined influence of LUCC, direct, and indirect climate change increased wheat yield by about 18.5% and 15.5% in the context of 1.5 and 2.0 °C warming, respectively. The relative contribution of LUCC, indirect climate change and direct climate change to yield was 4.7%, 41.2%, and 54.1% under 1.5 °C warming, and 13.1%, 28.7%, and 58.2% under 2.0 °C warming, respectively. We suggest that changes in irrigation water availability should be considered from a watershed perspective when simulating the influence of climate change on crop yield, especially regional crop production estimation. © 2023 Society of Chemical Industry.

Allocating Robots/Cobots to Production Systems for Productivity and Ergonomics Optimization

Collaboration between humans and robots has great promise in manufacturing systems. The utilization of cobots in a manufacturing system can improve both productivity and ergonomics. In this paper, we study the problem of how to allocate limited cobot/robots to manufacturing systems with multiple workstations so that an integrated performance measure, considering both productivity and ergonomics is optimized. Previous work on cobot/robot allocation in manufacturing systems focus on the decomposition of tasks for a single workstation into multiple work elements, and then split them between human and robots, rather than studying multi-machine systems. To bridge this gap, we consider the allocation of cobots/robots to a multi-stage manufacturing system. Specifically, we establish an integrated performance measure and formulate cobot/robot allocation into a constraint integer programming problem. With this formulation, we obtain the optimal allocation of one available cobot/robot in simulated production systems, based on the integrated performance measure of productivity and ergonomics. Furthermore, the allocation problems of production systems with multiple cobots/robots is considered and solved with a scalable algorithm. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Collaborative robots and independent robots are increasingly applied to manufacturing production systems. However, how to optimally allocate both types of robots considering both productivity and ergonomics influence has not been well studied. In this article, we established a practical optimization method to allocate cobots/robots to different workstations and split the work between cobot and human in one workstation when there are multiple workstations and a limited number of available cobots/robots in the manufacturing systems. Based on various real-world scenarios, we inferred useful insights for the robot/cobot allocation problem. To deal with the computational load when the number of workstations is large, a scalable optimization algorithm is also adopted. The case study results demonstrated the effectiveness of the proposed approach.

The Role and Development of Industrial Internet Platforms in the Digital Transformation of the Tobacco Industry

Abstract The tobacco industry, akin to traditional manufacturing, exhibits a high level of integration. The tobacco industrial Internet platform combines information from the tobacco industry, is able to efficiently interact with and process data, and plays a vital role in the transformation and development of the tobacco industry’s digitalization. Based on an analysis of the current state of digital transformation in the tobacco industry, we designed the tobacco industrial Internet platform. This platform primarily relies on the core of the information physical system (CPS), the industrial Internet of Things, and the production and manufacturing simulation system. It offers functions such as production preview, real-time simulation, and comprehensive life cycle management of production equipment. After putting the industrial Internet platform designed in this paper to the test in the tobacco industry, the amount of sewage from the tobacco boiler during production went down by 0.32 tons, and the amount of energy used to make cigarettes went down by 5.07 m³/box from before the platform was used to after it was used. This shows that the tobacco industrial Internet platform proposed in this paper is a good way to improve quality and efficiency, cut down on energy use, and improve the production of tobacco. This demonstrates that the implementation of the tobacco industrial Internet platform in this paper significantly enhances quality, reduces consumption, and conserves energy in tobacco production.

Toward data-driven and multi-scale modeling for production system simulation : Comparison of modeling methods

Toward data-driven and multi-scale modeling for production system simulation : Comparison of modeling methods

Unveiling the resilience of smallholder farmers in Senegal amidst extreme climate conditions

AbstractIn Senegal, agriculture is an important sector underpinning the socioeconomic fabric of the populace. Notably, the agricultural production in this region exhibits heightened sensitivity to climatic perturbations, particularly droughts and heat waves. This study aims to determine the resilience of different agronomic interventions for farmers practicing mixed farming that produce both crops (i.e., groundnut (Arachis hypogaea L.) and pearl millet (Pennisetum glaucum (L.) R. Br.)) and raise animals in the Groundnut Basin in Senegal, which holds historical and socioeconomic significance. To understand the current situation regarding demographics, economics, consumption behavior, and farm operations for smallholder farmers, data were comprehensively collected from government and nongovernment organizations (NGO) reports, scientific papers, organization databases, and surveys. Additionally, the Agricultural Production Systems sIMulator (APSIM) was used to understand how combinations of three planting dates, three plant densities, and six urea nitrogen (N) fertilizer rates affected the yield of pearl millet, which were used as the alternative scenarios to the baseline in the farm modeling and analyses. All the collected and generated data were used as inputs into the Farm Simulation Model (FARMSIM) to generate economic, nutritional, and risk data associated with mixed farming systems. The generated data were then used to determine the resilience of the alternative scenarios against the baseline. Initially, a multi‐objective optimization was employed to meet nutritional needs while maintaining a healthy diet at the lowest cost. Then, the scenarios that met the population's nutritional requirements were evaluated based on four economic indicators: net cash farm income (NCFI), ending cash reserves (EC), net present value (NPV), and internal rate of return (IRR). Lastly, those that passed the economic feasibility test were ranked based on risk criteria certainty equivalent (CE) and risk premium (RP). The analyses found N fertilizer rates of 0, 20, and 100 kg N ha−1 were generally economically not feasible. Additionally, medium (early‐July to late‐August) and late (late‐July to mid‐September) planting dates generally performed better than early (early‐June to late‐July) planting dates, while plant densities of 3.3 and 6.6 pL m−2 performed better than 1.1. The robust resilience approach introduced in this study is easily transferable to other regions.

Profit comparison analysis in production system simulation based on lean principles to achieve sustainable manufacturing

Companies can achieve effective and efficient process if they make continuous improvements to achieve sustainability. In general, companies are aware of waste in the production process, but do not carry out measurements and analyzes related to this waste, including unit cost and profit analysis. A production system is needed that is able to minimize the unit costs and maximize profits in the company, one of the concepts used is the lean concept. This study aims to analyze the cost unit and profit generated in production systems that apply lean principles. Production system simulation is carried out using the concepts of Heijunka, Jidouka, and Kanban System. These three concepts are applied to a production system simulation that uses miniature cars with unit cost and profit comparison outputs with traditional production systems, pull systems, heijunka, jidouka, and Kanban systems. The results show that the unit cost of simulations 1 to 6 is getting lower, while the profit is increasing. In simulations 1 and 2 no profit was obtained because of implementing the traditional system while in simulations 3 to 6 there was an increase in profit because they had applied lean principles to the production system. Simulations that have implemented the lean concept have low unit costs and increasing profits, but what distinguishes the work methods applied. Production systems that apply the lean concept can help a company achieve sustainability in the economic field.

Comprehensive process simulation of a biomass-based hydrogen production system through gasification within the BECCS concept in a commercial two-stage fixed bed gasifier

Comprehensive process simulation of a biomass-based hydrogen production system through gasification within the BECCS concept in a commercial two-stage fixed bed gasifier

Energy Analysis of an Existing CO2 Plant Simulated in Aspen Plus V8.8

Assessing energy consumption of a CO2 plant helps estimate the plant’s greenhouse gas emissions and its environmental impact, which then facilitate the development of an eco-friendly strategy. Main objective of this research is to conduct a fundamental energy analysis in Aspen Plus V8.8 for an already simulated CO2 production system utilizing natural gas as feed. The findings show that the available energy savings of the plant is 4.329×105 watt. There are no CO2 emissions, portraying a carbon-free synthesis process which is often the desire of environmental protection agencies. This analysis should be used in future studies to determine the appropriate sizing of equipment and utilities, ensuring that the plant operates at its optimal capacity and avoids overdesign.

Automated model generation framework for material flow simulations of production systems

Owing to increasingly important drivers for change, such as automation, digitisation, and dynamic demand patterns, simulation models of production systems become outdated rapidly. At the same time, building simulation models often requires much time, cost, and expertise, especially when dealing with complex job shop production systems. To address these challenges, an automated simulation model generation (ASMG) framework for material flow simulation of production systems is presented. This framework contains multiple approaches to infer routeing, control and temporal aspects from event-based data. To achieve this, methods from process mining (PM) and machine learning (ML) are applied. Additionally, the suitability of Coloured Petri Nets (CPNs) to serve as conceptual and operational simulation models is examined. The inferred simulation models have high validity when compared to the real system concerning the KPIs machine utilisation, throughput, and work in process. It is shown, that most model elements can be inferred very well, in particular process routes, processing times, and resource selection rules. This proof of concept serves as a foundation for research on detection approaches inferring further model elements such as setup times accurately.

Effects of climate factors on wheat and maize under different crop rotations and irrigation strategies in the North China Plain

The North China Plain (NCP) is a crucial agricultural region for grain production in China, primarily focusing on wheat and maize cultivation. However, these crop yields are highly vulnerable to fluctuations in temperature and precipitation, with climate change being a significant factor. This study investigates the impact of climate factors on wheat and maize yields in the NCP under various crop rotations and water supply strategies. Using the Agricultural Production Systems sIMulator crop mechanism model, we evaluated the effects of different crop rotations and water supply strategies on wheat and maize yields. A comprehensive analysis of the simulated results determined the yield variation trends and their correlation and sensitivity to different climate factors. The findings revealed that precipitation levels over the past 40 years showed no significant trend, while radiation levels showed a significant decreasing trend, and annual mean maximum and minimum temperatures displayed significant increasing trends. Furthermore, the study found that irrigation practices and crop rotations substantially impact grain yield in the study area, with average yields ranging from 8105.5 kg ha−1 under rainfed conditions to 13 088.8 kg ha−1 under fully irrigated conditions. There was a statistically significant trend of increasing yields for fully irrigated Monocrop-Wheat and decreasing yields for fully irrigated Monocrop-Maize over the past 40 years. Sensitivity analysis also showed that rational rotation and irrigation can reduce grain yield sensitivity to climate change. In conclusion, the prudent use of rotation and irrigation can enhance food production resilience to climate change. These findings have significant implications for developing strategies to optimize crop yields and adapt to climate change in the NCP while considering trade-offs.

Component-wise iterative ensemble Kalman inversion for static Bayesian models with unknown measurement error covariance

The ensemble Kalman filter (EnKF) is a Monte Carlo approximation of the Kalman filter for high dimensional linear Gaussian state space models. EnKF methods have also been developed for parameter inference of static Bayesian models with a Gaussian likelihood, in a way that is analogous to likelihood tempering sequential Monte Carlo (SMC). These methods are commonly referred to as ensemble Kalman inversion (EKI). Unlike SMC, the inference from EKI is asymptotically biased if the likelihood is non-linear and/or non-Gaussian and if the priors are non-Gaussian. However, it is significantly faster to run. Currently, a large limitation of EKI methods is that the covariance of the measurement error is assumed to be fully known. We develop a new method, which we call component-wise iterative EKI (CW-IEKI), that allows elements of the covariance matrix to be inferred alongside the model parameters at negligible extra cost. This novel method is compared to SMC on a linear Gaussian example as well as four examples with non-linear dynamics (i.e. non-linear function of the model parameters). The non-linear examples include a set of population models applied to synthetic data, a model of nitrogen mineralisation in soil that is based on the Agricultural Production Systems Simulator, a model predicting seagrass decline due to stress from water temperature and light, and a model predicting coral calcification rates. On our examples, we find that CW-IEKI has relatively similar predictive performance to SMC, albeit with greater uncertainty, and it has a significantly faster run time.

Global sensitivity analysis of APSIM-wheat yield predictions to model parameters and inputs

The performance of cropping system models is often limited by uncertainties in model parameters and inputs. This study aims to explore how model yield prediction responds to these uncertainties under various environmental and management conditions. The Sobol’ method was used to investigate the sensitivity of the Agricultural Production Systems SIMulator (APSIM)-Wheat yield prediction to the following factors: air temperature (maximum and minimum), precipitation, initial soil nitrogen content, nitrogen fertilization amount, and soil hydraulic parameters. Eighteen scenarios were defined to consider a combination of three weather conditions (wet, normal, and dry growing season, based on the historical climatology of the Wimmera district of Victoria, Australia), three soil types (sandy, loamy, and clayey soils), and two nitrogen fertilization applications (50 and 100 kg N/ha). Eight thousand APSIM simulations were used to calculate the first order and total sensitivity indices for each scenario. The effects of weather, soil water characteristics and nitrogen availability were estimated by measuring their impacts on sensitivity indices. Our results show that yield was more sensitive either to variables that control water availability (precipitation and soil type) or variables that control nitrogen availability, depending on which was more limiting to wheat growth under the scenario. For example, in case of low nitrogen fertilization scenario, the initial nitrogen content sensitivity ranked first. Variation of this factor contributed 64% to 79% to the variance in simulated yield for clayey soils, nearly four times higher than the second-ranked factor (nitrogen fertilization amount). Soil hydraulic parameters and precipitation were most important when the crop growth was more constrained by water availability than by nitrogen availability. In the case of sandy soils in dry years with high nitrogen fertilization level, soil parameters and precipitation accounted for 83% of the yield variability. Minimum temperature consistently ranked as the least important factor under all scenarios. This work will help researchers better understand the model inputs’ impacts on the simulated yield variability under various soil, management, and weather conditions. The results also support agronomists and practitioners in the optimal use of the APSIM-Wheat model as a management tool.

I-RAT: A discussion support system to rapidly assess economic and environmental impacts of different sugarcane irrigation practices

Water pollution and climate change are among the greatest threats to the iconic Great Barrier Reef (GBR). To improve the GBR’s long-term outlook, making improvements on these fronts is required. Owing to the complex interactions between soils, climate, and farm management, a tool is needed to guide agricultural land managers about which management practice changes are more likely to deliver improved water quality and climate resilience outcomes whilst maintaining profitability. Using concepts drawn from social studies of science and technology, a ‘discussion’ support system (DSS) named Irrigation Rapid Assessment Tool (i-RAT) was developed through participatory processes designed to enhance co-learning from the development of this DSS. i-RAT is a rapid assessment-visualisation tool in an interactive web application (https://i-rat.net) to visualise the impacts of changed irrigation practices for sugarcane farmers and extension staff. Specifically, i-RAT compares these impacts regarding farm economics, water quality, productivity, carbon and nitrogen cycles, and greenhouse gas emission. The core of i-RAT is the sugarcane module from the Agricultural Production Systems sIMulator (APSIM) modelling platform. i-RAT was first developed for the Burdekin sugarcane growing region in Queensland (Australia’s largest sugarcane producing region). Various soil types and management scenarios representing farming practices in the Burdekin region were used for simulations (1971–2021) to generate a datacube. In this paper, we describe how fundamental learnings about more (drought) resilient farming systems and more sustainable irrigation practices can be extracted from i-RAT. Details of the participatory approach with research and industry partners that informed the design and function of i-RAT and how APSIM was set up and parameterised are described. A ‘what-if’ analysis demonstrated the i-RAT features and application.

Enhancing gas production through advanced detailed design engineering: a global perspective on industry practices

Enhancing gas production through advanced detailed design engineering has become a pivotal approach in the global energy sector. This paper explores the integration of cutting-edge engineering methodologies to optimize gas production processes, reduce operational inefficiencies, and meet increasing energy demands. The application of detailed design engineering in gas production involves sophisticated modeling, simulation, and analysis of production systems, ensuring enhanced precision, safety, and scalability. By leveraging advanced computational tools, engineers can predict and mitigate potential production bottlenecks, optimize well designs, and implement efficient gas processing techniques. The global perspective on industry practices highlights regional variations in design standards, regulatory frameworks, and technological adoption. For example, North American gas fields utilize integrated digital twins and automation technologies to streamline production, while regions like the Middle East emphasize design optimization in offshore gas extraction. Additionally, the development of sustainable practices, such as carbon capture and storage (CCS) and energy-efficient processing units, is becoming a critical focus for gas producers striving to reduce environmental impact and align with international climate goals. This study also examines the role of collaborative frameworks between engineering teams and production operators in fostering innovation and knowledge transfer, further driving the refinement of detailed design approaches. The future of gas production will depend significantly on advancements in materials science, data analytics, and artificial intelligence, which are expected to revolutionize the design and operation of gas production systems. In conclusion, enhancing gas production through advanced detailed design engineering offers the potential to unlock new efficiencies, improve safety standards, and contribute to global energy security. However, the successful implementation of these practices requires continuous investment in research, technology, and cross-industry collaboration.

Analysis of metaheuristic optimisation techniques for simulated matrix production systems

Increasing demand for individualised products has led to the concept of mass customisation, combining high product variety with production efficiency coming along with mass production. Companies are moving to matrix production systems with complex product flows for mass customisation. One challenge in such systems is the determination of optimal system configurations to fulfil future demands while minimising production costs. An approach to determine the ideal configuration is to use metaheuristics like genetic algorithms or simulated annealing to optimise simulation models. However, it is unclear which methods are ideally suited to finding the best solutions. This contribution compares the performance of genetic algorithms and simulated annealing when optimising the configuration of a company-specific matrix production system using discrete event simulation. The methods are evaluated using different objective functions. For the genetic algorithm, different observation strategies are also tested. Overall, the simulated annealing approach delivers better results with shorter solution times. The contributing factors leading to the different results are discussed, and areas for future research are pointed out.

Climate Change and Soil Dynamics: A Crop Modelling Approach

The impact of global climate change is a challenge to the sustainability of many ecosystems, including soil systems. However, the performance of soil properties under future climate was rarely assessed. Therefore, this study was carried out to evaluate selected soil processes under climate change using an agri-environmental modeling approach to Sri Lanka. The Agricultural Production Systems Simulator (APSIM) model was used to simulate soil and plant-related processes using recent past (1990–2019) and future (2041–2070) climates. Future climate data were obtained for a regional climate model (RCM) under representative concentrations pathway 4.5 scenarios. Rainfalls are going to be decreased in all the tested locations under future climate scenarios while the maximum temperature showcased rises. According to simulated results, the average yield reduction under climate change was 7.4%. The simulated nitrogen content in the storage organs of paddy declined in the locations (by 6.4–25.5%) as a reason for climate change. In general, extractable soil water relative to the permanent wilting point (total available water), infiltration, and biomass carbon lost to the atmosphere decreased while soil temperature increased in the future climate. This modeling approach provides a primary-level prediction of soil dynamics under climate change, which needs to be tested using fieldwork.

Genotype × environment × agronomic management interaction to enhance wheat yield in the Mediterranean rainfed environments of Morocco: II. Process based modeling

Durum wheat (Triticum turgidum subsp. durum) is the oldest and most cultivated cereal crop in Middle East and North Africa (MENA) region and under Mediterranean climatic conditions. Morocco is one of the largest producer of durum wheat in MENA region, cultivated in more than 1 million ha area produced 2.5 million tons in 2020, which accounts for 17% of the total production in the region. In the region, rainfed production system is predominant, and with declining rainfall amounts with high variability, increasing water scarcity, and suboptimal input application, its productivity growth is low and needs to be increased to fulfill the growing demand. Developing context-specific management advisory is needed to improve productivity and resilience under such variable rainfed production environments. Agricultural Production Systems sIMulator (APSIM) model was calibrated and evaluated using four years (2015–2019) of on-station experimental data from genotype × seeding time × water management experiment conducted at International Center for Agricultural Research in the Dry Areas (ICARDA) research station, Morocco. Long-term (1984–2021) simulation was carried out to determine the contribution of Genotype × Environment × Management components for sustainably improving crop productivity. The results showed rainfall or supplementary irrigation (23–36%) followed by N fertilizer (28–38%), cultivar (9–14%), and seeding date (7–14%) have the largest contribution to the yield variance of durum wheat in Merchouch, Meknes, and Sidi El Aidi regions of Morocco. Under rainfed conditions, wheat yield was highest in Merchouch (4.5 t ha−1) and lowest (1.8 t ha−1) in Sidi El Aidi. Due to significant rainfall variability, the seeding date varies across year and location; however, generally, it is between 2nd week of November to 1st week of December. Under rainfed conditions, seeding after 1st week of December caused the average yield reduction of 120, 81, and 31 kg ha−1 d−1 in Merchouch, Meknes, and Sidi El Aidi, respectively. In all locations, short-duration varieties provided higher averaged yields with better resilience than medium and long-duration varieties. Decomposing yield variance caused by Genotype × Environment × Management provides the opportunity for risk reduction, improvement of wheat yield and resilience, and designing climate-smart adaptation strategies in rainfed Mediterranean conditions. Our findings highlight one-size-fits-all approach is inadequate and context-specific tailored agronomic practices and suitable genotypes is crucial for achieving sustainability and resilience of wheat production in variable climatic condition in Morocco and similar production environment.

Modeling the effect of soil fertility management options on maize yield stability under variable climate in a sub-humid zone in Ghana

IntroductionUncertainty in the yield of maize due to variability in weather is a major challenge to smallholder farmers in Sub Sahara Africa. This study explores the potential of combining locally available organic resources and inorganic fertilizer to increase grain yield and reduce variability in yields associated with variations in rainfall distribution.MethodsTo assess the effectiveness of this practice, the Agricultural Production Systems sIMulator (APSIM) crop model was calibrated and evaluated using maize experiments on nutrient management options. The evaluated model was used to simulate maize growth and yield using multiple-year data (1984–2018) under different planting windows for two growing seasons. The treatments were (i) control, (ii) inorganic fertilizer alone (INOFRecom), combining organic resources [empty fruit bunch of palm (EFB) and compost with inorganic fertilizer (INOF) to make up equivalent nutrients in (ii)]; (iii) EFB + INOF and (iv) Comp + INOF.ResultsThough all the soil amendments boosted grain output, the EFB + INOF treatment outperformed the other treatments in the major season with gains of between 161 and 211% and the most stable (least inter-annual variability of 27%) yield. Across the planting windows, the INOFRecom and EFB + INOF treatments achieved comparable yield increments in the minor season. Though grain yield variability was high during the minor rainy season, combining organic and inorganic fertilizers reduced inter-annual yield variability, thus, lowering uncertainty in yield due to variable inter-annual rainfall. Combining local organic resources with a reduced amount of inorganic fertilizer produced higher yields and better yield stability compared to using only inorganic fertilizer.DiscussionsThus, such soil fertility management solutions might sustain resource use and boost maize grain yield in the study area, where strategies for sustainable crop nutrition remain a critical necessity. The improved nitrogen management regimes may result in fewer environmental hazards for vulnerable rainfed agricultural systems.

Predicting maize yield in Northeast China by a hybrid approach combining biophysical modelling and machine learning

ContextChina produces more than 20 % of maize grain in the world, and Northeast China (NEC) accounts for ∼30 % of the nation's total maize production. Previous studies have used either climate data, satellite data, or crop growth model (CGM) to predict or forecast maize yield. However, maize is highly susceptible to the effect of extreme climate events (such as drought, heat) in NEC, and there is a lack of studies to predict/forecast maize yield by integrating climate data, satellite data, extreme climate events, and CGM-simulated data. ObjectiveWe aim to develop a hybrid approach with machine learning to blend different sources of data (climate data, satellite data, extreme climate events) and process-based modelling results to improve predictive accuracy of maize yield in NEC. MethodsUsing maize data from 44 sites during the period of 2000–2013 in NEC, we firstly optimized Agricultural Production System sIMulator (APSIM) using Differential Evolution Adaptive Metropolis combined with Gaussian likelihood function and Bayesian multiplication method. Next, we divided the growing season into five phases, and selected variables of different phases using exploratory data analysis and Random Forest. Then, we developed a hybrid model using Random Forest by blending of multiple sources of data and APSIM simulations to predict maize yield from the start to the end of the growing season, and quantified the relative contribution of predictors. Results and conclusionsA hybrid model developed with random forest by combining climate data, NDVI, extreme climate events and APSIM simulations can achieve high performance for predicting yield toward the end of the growing season. The accuracy of in-season yield prediction showed a linear increase with MAPE/KGE changing from 19 %/0.05 to 13 %/0.53 from start to end of the growing season. Yield forecasts are acceptable with RMSE/MAE of 1.20/1.01 Mg ha−1 (16 %/13 % of the observed mean yield) approximately one-month prior to harvest. The most important predictor that affect yield forecast was APSIM-simulated biomass or yield, and the most important extreme climate event was drought during early grain-filling stage. SignificanceWith the increasing availability of crop-related data, we expect that the in-season forecasting capacity of the proposed methodology could be further improved, and the methodology can be extended to other crops and other regions for yield forecast.