Operations research and machine learning to manage risk and optimize production practices in agriculture: good and bad experience

Cock, James H.; Jiménez, Daniel; Dorado, Hugo; Oberthür, Thomas

doi:10.1016/j.cosust.2023.101278

Cited by 8 publications

(6 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, previous yield gap analyses in SSA are narrow in geographic scope and range of agronomic practices 8 – 10 , 35 , 40 , 41 . In contrast, our study includes thousands of farmers from seven countries, across 25 production environments, including farmers with access to inputs which, in turn, extend the range of agronomic practices in our database, increasing the odds to identify yield-improving practices 42 . Finally, the combination of a climate spatial framework, and statistical and machine-learning methods, together with soil and terrain databases, allowed us to control for the confounding effects of varying climate, soil, and terrain factors in relation to the impact of agronomic practices on on-farm yields.…”

Section: Discussionmentioning

confidence: 99%

“…About half of the fields in the database comprised farmers who subscribed to the One Acre Fund program whereas the other half did not. We see the inclusion of farmers with varying levels of technology adoption in the database as an advantage as it allows to increase the variation in management practices across farmer fields 42 . Because farmers engaged with the One Acre Fund program have greater technology adoption, it was not surprising that the average maize yield of our database (3 t ha − 1 ) was higher than the average maize yield in those same regions (1.7 t ha − 1 ) 1 .…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Adopting yield-improving practices to meet maize demand in Sub-Saharan Africa without cropland expansion

Aramburu-Merlos,

Tenorio,

Mashingaidze

et al. 2024

Nat Commun

View full text Add to dashboard Cite

Maize demand in Sub-Saharan Africa is expected to increase 2.3 times during the next 30 years driven by demographic and dietary changes. Over the past two decades, the area cropped with maize has expanded by 17 million hectares in the region, with limited yield increase. Following this trend could potentially result in further maize cropland expansion and the need for imports to satisfy domestic demand. Here, we use data collected from 14,773 smallholder fields in the region to identify agronomic practices that can improve farm yield gains. We find that agronomic practices related to cultivar selection, and nutrient, pest, and crop management can double on-farm yields and provide an additional 82 million tons of maize within current cropped area. Research and development investments should be oriented towards agricultural practices with proven capacity to raise maize yields in the region.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Adopting yield-improving practices to meet maize demand in Sub-Saharan Africa without cropland expansion

Aramburu-Merlos,

Tenorio,

Mashingaidze

et al. 2024

Nat Commun

View full text Add to dashboard Cite

show abstract

“…OR looks for better ways to conduct organizational operations using mathematical, computer-based, or other analytical methods [95]. OR has been divided into two branches, hard OR and soft OR.…”

Section: Discussionmentioning

confidence: 99%

Solving the Problem of Reducing the Audiences’ Favor toward an Educational Institution by Using a Combination of Hard and Soft Operations Research Approaches

Xu,

Edalatpanah,

Sorourkhah

2023

Mathematics

View full text Add to dashboard Cite

Because of hyper-complexity, a difficulty to define, multiple stakeholders with conflicting perspectives, and a lack of clear-cut solutions, wicked problems necessitate innovative and adaptive strategies. Operations research (OR) has been a valuable tool for managers to make informed decisions for years. However, as we face increasingly complex and messy problems, it has become apparent that relying solely on either hard or soft OR approaches is no longer sufficient. We need to explore more innovative methodologies to address these wicked problems effectively. This study has bridged the research gap by proposing a structured process encompassing a subdivision-based problem structuring method for defining the wicked problem, a multi-attribute decision-making (MADM) for prioritizing subproblems, and a hard OR technique, data envelopment analysis (DEA) for tackling one of the most critical subdivisions. The proposed methodology, the subdivision-based problem structuring method (SPSM), implemented in a case study, focuses on a higher education institution experiencing a decline in student admissions and involves five steps. First, a diverse group of stakeholders is formed to ensure the comprehensive consideration of perspectives. Second, the wicked problem is defined, considering long-term consequences, multiple stakeholders, and qualitative stakeholder opinions. Third, a hierarchical structure is created to break down the wicked problem into manageable subproblems. Fourth, a multi-criteria decision-making (MCDM) method prioritizes subproblems. Finally, the subproblems are addressed one by one using a combination of soft and hard OR tools. The findings highlight the benefits of integrating hard and soft OR approaches. The study concludes with reflections on the implications of using a combined OR approach to tackle wicked problems in higher education and beyond.

show abstract

“…Second, previous yield gap analyses in SSA are narrow in geographic scope and range of agronomic practices [38][39][40][41] . In contrast, our study includes thousands of farmers from four countries, across 21 climate zones, including farmers with access to inputs which, in turn, extend the range of agronomic practices in our database, increasing the odds to identify yield-improving practices 42 . Finally, the combination of a robust climate spatial framework and statistical methods, together with soil and terrain databases, allowed us to control for the confounding effects of varying climate, soil, and terrain factors in relation to the impact of agronomic practices on on-farm yields 7-9, 43,44 .…”

Section: Discussionmentioning

confidence: 99%

“…About half of the elds in the database comprised farmers who subscribed to the One Acre Fund program and the other half farmers who did not. We see inclusion of farmers with varying level of technology adoption in the database as an advantage as it allows to increase the variation in management practices across farmer elds 42 . Because farmers engaged with the One Acre Fund program have greater technology adoption, it was not surprising that the average maize yield of our database (2.9 t ha − 1 ) was higher than the average maize yield in those same regions (1.6 t ha − 1 ) 1 .…”

Section: Database Descriptionmentioning

confidence: 99%

Sub-Saharan Africa needs better agronomy to meet maize demand without massive cropland expansion and imports

Grassini,

Aramburu-Merlos,

Tenorio

et al. 2023

Preprint

View full text Add to dashboard Cite

Maize demand in sub-Saharan Africa will increase 2.3 times during the next 30 years driven by demographic and dietary changes. Over the past two decades, the area cropped with maize has expanded by 17 million hectares in the region, with little improvement in farmer yields. If these trends persist, massive increase in maize area and/or imports will be needed to meet domestic demand. Here we used data collected from 13,364 smallholder fields to identify agronomic practices that can promptly deliver large on-farm yield gains. We show that good agronomic practices regarding nutrient, crop management, and cultivar selection can more than double on-farm yields and provide an additional 93 million tons of maize within current cropland. Given the magnitude and urgency of the challenge ahead, agricultural research and development investments must be re-oriented towards agricultural practices with proven capacity to raise yields in farmer fields.

show abstract

Operations research and machine learning to manage risk and optimize production practices in agriculture: good and bad experience

Cited by 8 publications

References 41 publications

Adopting yield-improving practices to meet maize demand in Sub-Saharan Africa without cropland expansion

Adopting yield-improving practices to meet maize demand in Sub-Saharan Africa without cropland expansion

Solving the Problem of Reducing the Audiences’ Favor toward an Educational Institution by Using a Combination of Hard and Soft Operations Research Approaches

Sub-Saharan Africa needs better agronomy to meet maize demand without massive cropland expansion and imports

Contact Info

Product

Resources

About