FIGURE 5 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
| Projected outcome distribution (top left), high-value variables (EVPI; top right), project cash flow (bottom left) and important variables (determined by PLS regression; bottom right) for the overall project in Witu, Kenya. Results were produced through Monte Carlo simulation (with 10,000 model runs) of project performance over 10 years. Red and green bars in the outcome distribution indicate positive and negative values, respectively. In the PLS plot, green bars indicate positive correlations of uncertain variables with the outcome variable, while red bars indicate negative correlations.
Source publication
p>Designing and implementing biodiversity-based value chains can be a complex undertaking, especially in places where outcomes are uncertain and risks of project failure and cost overruns are high. We used the Stochastic Impact Evaluation (SIE) approach to guide the Intergovernmental Authority on Development (IGAD) on viable investment options in h...
Context in source publication
Context 1
... both positive and negative farmer outcomes appeared plausible according to the simulation results, the EVPI analysis indicated that additional information about four variables could potentially change the recommendation emerging from the SIE process. These were the honey price, with a value of information of about Ksh 2750, followed by honey production per hive (Ksh The distribution of the projected NPV for the project in Witu had a median of Ksh 36 million, with 90% confidence that the actual NPV for the project lay within the range of Ksh-14-150 million. The model responded most sensitively to the amount of honey produced per hive. Seven other variables also had important impact on projected outcome values ( Figure ...
Similar publications
Decisions to invest in alternative intelligent transportation system (ITS) technologies are expected to increase in complexity, particularly with the introduction of connected vehicles (CV) and automated vehicles (AV) in the coming years. Traditional alternative analyses based on deterministic return on investment analysis are unable to capture the...
Citations
... The procedures applied were derived from decision analysis approaches [31,44,55,56] and combined with the adjusted procedures of Vieux et al. [16]. A literature review for animal-based product GHGEs was conducted to create value ranges. ...
... For model implementation, an uncertainty analysis via MC simulation was conducted, combining approaches from Vieux et al. [16] and decision analyses [31,46,55,56]. MC simulation has been frequently applied to quantify uncertainty in environmental contexts and complex natural systems [27,31]. ...
... This means that even with better data, diets with large amounts of animal-based products are expected obstacles to achieving climate mitigation goals (even if for example population growth is lower than expected or a better representation of the actual products would lead to lower CFs). [31,46,55,56]), using additional data and re-and add to uncertainty, particularly if only a few sources, but also when many data points are available. The data can also (b) be outdated in some cases, and thus do not account for improvements over time (e.g., increased input efficiency) different GHGE data sources can augment uncertainty and variability, but also increase the chance of containing the true value in the final result. ...
The food system plays a crucial role in mitigating climate change. Even if fossil fuel emissions are halted immediately, current trends in global food systems may prevent the achieving of the Paris Agreement’s climate targets. The high degree of variability and uncertainty involved in calculating diet-related greenhouse gas emissions limits the ability to evaluate reduction potentials to remain below a global warming of 1.5 or 2 degrees. This study assessed Western European dietary patterns while accounting for uncertainty and variability. An extensive literature review provided value ranges for climate impacts of animal-based foods to conduct an uncertainty analysis via Monte Carlo simulation. The resulting carbon footprints were assessed against food system-specific greenhouse gas emission thresholds. The range and absolute value of a diet carbon footprint become larger the higher the amount of products with highly varying emission values in the diet. All dietary pattern carbon footprints overshoot the 1.5 degrees threshold. The vegan, vegetarian, and diet with low animal-based food intake were predominantly below the 2 degrees threshold. Omnivorous diets with more animal-based product content trespassed them. Reducing animal-based foods is a powerful strategy to decrease emissions. However, further mitigation strategies are required to achieve climate goals.
... Involving experts through collaborative modeling approaches is crucial for development of credible decision support systems, and it allows accounting for agricultural risks (Oliver et al., 2012). Probabilistic model simulations using collaborative workshops to generate input data based on expert estimates have been applied in several studies to support stakeholders' risk management in production systems around the globe (Do et al., 2020;Rosenstock et al., 2014;Tamba et al., 2017;Wafula et al., 2018;Yigzaw et al., 2019). In horticultural production, similar approaches have been applied to predict the outcome of risk management strategies in sweet cherry production (Rojas et al., 2021). ...
Many farmers hesitate to adopt new management strategies with actual or perceived risks and uncertainties. Especially in ornamental plant production, farmers often stick to current production strategies to avoid the risk of economically harmful plant losses, even though they may recognize the need to optimize farm management. This work focused on the economically important and little-researched production system of ornamental heather (Calluna vulgaris) to help farmers find appropriate measures to sustainably improve resource use, plant quality, and profitability despite existing risks. Probabilistic cost-benefit analysis was applied to simulate alternative disease monitoring strategies. The outcomes for more intensive visual monitoring, as well as sensor-based monitoring using hyperspectral imaging were simulated. Based on the results of the probabilistic cost-benefit analysis, the expected utility of the alternative strategies was assessed as a function of the farmer's level of risk aversion. The analysis of expected utility indicated that heather production is generally risky. Concerning the alternative strategies, more intensive visual monitoring provides the highest utility for farmers for almost all levels of risk aversion compared to all other strategies. Results of the probabilistic cost-benefit analysis indicated that more intensive visual monitoring increases net benefits in 68% of the simulated cases. The application of sensor-based monitoring leads to negative economic outcomes in 85% of the simulated cases. This research approach is widely applicable to predict the impacts of new management strategies in precision agriculture. The methodology can be used to provide farmers in other data-scarce production systems with concrete recommendations that account for uncertainties and risks.
Supplementary information:
The online version contains supplementary material available at 10.1007/s11119-022-09909-z.
... Decision analysis procedures would thus greatly benefit from improved mechanisms to identify a preferable option even in a context of high residual uncertainty. So far, this initial call has usually been based on the concept of expected value of the outcome variable (Luedeling et al. 2015;Wafula et al. 2018;Whitney et al. 2017;Ruett et al. 2020), but this concept is ignorant of decision makers' attitude toward risk, which features prominently in many decision-making processes. To add nuance to the initial evaluation on which decision option should be preferred, we ranked the outcomes of the strategies using stochastic dominance. ...
Decision making in pest management is a challenging task. While pest dynamics are often quite uncertain, such decisions are often based on tenuous assumptions of certainty (economic injury levels and marginal utility approximations). To overcome such assumptions and adequately consider uncertainty, we apply decision analysis to evaluate management strategies used by farmers in the Colombian Caribbean against the boll weevil (BW). We represent the decision to protect the crop using partial budget analysis. This allows us to capture key properties of BW control strategies, while accounting for uncertainty about pest infestation pressure, control effectiveness and cotton yield and price. Our results indicate that proactive pest management is more efficient than reactive control given the current BW infestation pressure. However, farmers may prefer the reactive strategy, since they have experienced seasons with low infestation pressure where no insecticide applications were required. The proactive strategy, in contrast, requires scheduled pesticide applications in all years. Results show that in seasons with high infestation pressure the expected revenues of the reactive strategy tend to decrease, mainly because more spray applications are required when fields are heavily infested by the weevil. Value of information analysis revealed that uncertainties related to the start of the infestation, loss damage rate and attainable yield have the greatest influence on the decision recommendation for crop protection. Narrowing these key knowledge gaps may offer additional clarity on the performance of the current management strategies and provide guidance for the development of strategies to reduce insecticide use. This is particularly important for the promotion of the proactive strategy, which, under the current infestation pressure, has potential to reduce insecticide use. While economic injury levels can only be applied to responsive measures, our approach of partial budget analysis under uncertainty allows us to assess and compare both responsive and preventive measures in the same methodological framework. This framework can be extended to non-pesticide control measures.
... Monte Carlo-based decision models (Favretto et al. 2017;Rosenstock et al. 2014) as well as Bayesian network models (Whitney et al. 2018) can help provide robust guidance for decision-making without cost-intensive long-term data collection. In the field of agriculture, Decision Analysis approaches have been used for many applications, e.g., to assess investment options in honey value chains (Wafula et al. 2018), the nutritional value of home gardens (Whitney et al. 2017), intervention options to prevent reservoir sedimentation (Lanzanova et al. 2019), or the viability of agroforestry investments (Do et al. 2020). ...
Voluntary standards can help to ensure the quality of projects eligible for carbon offsetting, i.e., selling carbon certificates. However, in deciding on whether to adopt such standards, the managers of carbon offset projects are faced with uncertainty regarding the costs and risks involved. Decision Analysis provides a helpful set of tools that can support such decisions by forecasting outcomes under different scenarios. We applied Decision Analysis methods to generate models for the decisions to certify two projects in Costa Rica with the voluntary carbon offset label “The Gold Standard”. We evaluated certifying an additional site of a partially certified reforestation project, as well as the initial certification of an agroforestry project.
We calibrated and interviewed decision-makers and stakeholders of the certification projects to identify important parameters and translate these into a decision model. We ran the final decision model as a Monte Carlo simulation to project plausible ranges of decision outcomes, expressed as Net Present Values and annual cash flows. We identified critical uncertainties and research priorities by using the Expected Value of Perfect Information. The results indicate that certification of the two projects would result in a positive Net Present Value. The partially low return on investment of the certification, however, shows the need for projects to undergo thorough evaluation and generate customized strategies before participating in a voluntary carbon offset scheme. The Decision Analysis approaches we describe can help to improve the process of decision making under uncertainty and should be widely adopted for evaluating the potential impacts of certification.
... Stochastic Impact Evaluation (SIE) is an ex-ante appraisal approach that allows integration of different impact dimensions by considering the intervention as a business case and by capturing the variability and uncertainty of all input variables Luedeling and Shepherd, 2016). SIE is a probabilistic decision modeling approach (Wafula et al., 2018), which is suitable for evaluating the outcomes of development projects, even when data availability is limited (cf. Luedeling et al., 2015;Rosenstock et al., 2014;Whitney et al., 2017). ...
... We used an expert knowledge elicitation approach Tamene et al., 2011;Whitney et al., 2018Whitney et al., , 2017 to collect estimates for model variables. All experts were familiarized with estimation techniques to overcome bias, allowing them to provide quantitative estimates in the form of distributions representing their subjective 90% confidence intervals for model variables Rosenstock et al., 2014;Wafula et al., 2018). Initial monetary estimates were collected in Ethiopian currency (i.e. ...
... These are the variables for which further measurement (reduction of uncertainty) could improve certainty about whether model outcomes are positive or negative . We used the expected value of perfect information (EVPI) procedure described by Wafula et al. (2018), to calculate the monetary value of additional information for the decisionmaking process. EVPI indicates the maximum value that a decision maker should be willing to pay in order to reduce uncertainty about which decision alternative promises the greatest returns (Whitney et al., 2017). ...
... Monte Carlo-based decision models (Favretto et al., 2017;Rosenstock et al., 2014) as well as Bayesian network models (Whitney et al., 2018) can help provide robust guidance for decision-making without cost-intensive long-term data collection. In the eld of agriculture, decision analysis approaches have been used for many application, e.g. to assess viable investment options in honey value chains (Wafula et al., 2018), the nutritional value of home gardens (Whitney et al., 2017), intervention options to prevent reservoir sedimentation (Lanzanova et al., 2019) or the viability of agroforestry investments (Do et al., 2020). ...
Voluntary standards help to ensure the quality of projects eligible for carbon offsetting, i.e. selling carbon certificates. However, in deciding on whether to adopt such standards the managers of carbon offset projects are faced with uncertainty regarding the costs and risks involved. Decision Analysis provides a helpful set of tools that can support such decisions by forecasting outcomes under different scenarios. We applied Decision Analysis methods to generate models for the decision to certify two projects in Costa Rica with the voluntary carbon offset label Gold Standard. We evaluated certifying an additional site of a partially certified reforestation project, as well as the initial certification of an agroforestry project.
We calibrated and interviewed decision-makers and stakeholders of the certification projects to identify important parameters and translated these into a decision model. We ran the final decision model as a Monte Carlo simulation to project plausible ranges of decision outcomes, expressed as Net Present Values and annual cash flows. We identified critical uncertainties and research priorities by using the Expected Value of Perfect Information. The results indicate that certification of the two projects would result in a positive Net Present Value. The partially low return on investment of the certification, however, shows the need for projects to undergo thorough evaluation and generate customized strategies before participating in a voluntary carbon offset scheme. The Decision Analysis approaches we describe can help to improve the process of decision making under uncertainty and should be widely adopted for evaluating the potential impacts of certification.
... We used the SIE framework ( Fig. 2), based on Luedeling and Shepherd (2016) to simulate FLR outcomes. SIE is a mixed methods approach that has been widely used to simulate outcomes under uncertainty and risk for investments in honey value chains (Wafula et al., 2018), water supply (Luedeling et al., 2015), irrigation development (Yigzaw et al., 2019), management of reservoir sedimentation (Lanzanova et al., 2019), and to determine the value of ecosystem services in rangelands (Favretto et al., 2017). In this study, we applied SIE as an iterative five-step process that supports decisions by integrating evidence and expert opinion in quantitative simulations of decision impact pathways (Fig. 2). ...
Forest and Landscape Restoration (FLR) is carried out with the objective of regaining ecological functions and enhancing human well-being through intervention in degrading ecosystems. However, uncertainties and risks related to FLR make it difficult to predict long-term outcomes and inform investment plans. We applied a Stochastic Impact Evaluation framework (SIE) to simulate returns on investment in the case of FLR interventions in a degraded dry Afromontane forest while accounting for uncertainties. We ran 10,000 iterations of a Monte Carlo simulation that projected FLR outcomes over a period of 25 years. Our simulations show that investments in assisted natural regeneration, enrichment planting, exclosure establishment and soil-water conservation structures all have a greater than 77% chance of positive returns. Sensitivity analysis of these outcomes indicated that the greatest threat to positive cashflows is the time required to achieve the targeted ecological outcomes. Value of Information (VOI) analysis indicated that the biggest priority for further measurement in this case is the maturity age of exclosures at which maximum biomass accumulation is achieved. The SIE framework was effective in providing forecasts of the distribution of outcomes and highlighting critical uncertainties where further measurements can help support decision-making. This approach can be useful for informing the management and planning of similar FLR interventions.
... Comparable results were presented in previous studies assessing a number of decision options in agricultural systems. For example, the most important variables for the success of biodiversity-based honey value chains in Kenya were honey price and the amount of honey obtained per hive (Wafula et al., 2018). Similarly, market price was also considered one of the most important variables in assessments of water management for farm communities in Africa (Hagos et al., 2012;Luedeling et al., 2015;Yigzaw et al., 2019). ...
Available options for mitigating the impacts of extreme weather events on temperate fruit trees involve a number of risks and uncertainties, leaving growers hesitant about the benefits of implementing new technologies to protect their orchards. We used Decision Analysis approaches, which account for these risks and uncertainties, to assess investment decisions in sweet cherry production systems in central Chile. We evaluated the adoption of polyethylene covers for orchards in northern-and southern-central Chile. Gathering expert and key stakeholder knowledge, we identified relevant variables for the adoption decision and developed a causal impact pathway model. We parameterized this model by collecting estimates from experts in the form of probability distributions. We implemented the model as a Monte Carlo simulation and projected probability distributions for the Net Present Value and the annual cash flow. Results highlight that farmers in southern-central Chile could expect major benefits from covering their orchards, with a 90% confidence interval for the Net Present Value from − 33,605 USD to 595,447 USD. In northern-central areas, implementing covers did not significantly improve the Net Present Value (90% confidence interval from − 149,597 USD to 433,361 USD). Across zones, our model results were sensitive to market price, crop yield, and fruit quality problems (i.e. low firmness). Cover effectiveness against rain events was relevant only in the southern-central zone, whereas effectiveness against frost events was relevant in both sites. Expected Value of Perfect Information calculations revealed that additional information on yield and market price could substantially help to make a confident decision. Our simulations suggest that orchard protection in southern-central Chile may be necessary for secure and profitable cherry production in the future. In this case study, we demonstrated the applicability of Decision Analysis to support farmers in identifying appropriate and effective strategies in time to overcome future challenges resulting from climate change.
... More recently, MC has also been used for evaluation of agricultural development investments. Wafula et al. [23] used MC to evaluate investment options in honey value chains in Kenya. Lanzanova et al. [24] used this approach to prioritize reservoir protection investments in Burkina Faso. ...
Agricultural development projects have a poor track record of success mainly due to risks and uncertainty involved in implementation. Cost-benefit analysis can help allocate resources more effectively, but scarcity of data and high uncertainty makes it difficult to use standard approaches. Bayesian Networks (BN) offer a suitable modelling technology for this domain as they can combine expert knowledge and data. This paper proposes a systematic methodology for creating a general BN model for evaluating agricultural development projects. Our approach adapts the BN model to specific projects by using systematic review of published evidence and relevant data repositories under the guidance of domain experts. We evaluate a large-scale agricultural investment in Africa to provide a proof of concept for this approach. The BN model provides decision support for project evaluation by predicting the value—measured as net present value and return on investment—of the project under different risk scenarios.
... These distributions can be developed using expert elicitation approaches (Bolger and Wright, 2017) for generating inputs for probabilistic simulations (Lanzanova et al., 2019), either alone or in combination with available data (Jochmann et al., 2010). Expert judgment has been shown to be highly useful for emulating decision situations (Wafula et al., 2018). Expert knowledge has been used as a data source in a diverse range of studies, such as risk assessment of supply chains (Giannakis and Papadopoulos, 2016), sustainability risk analysis (Valinejad and Rahmani, 2018) and life cycle assessment (Florindo et al., 2020). ...
Agricultural management decisions are usually made without perfect knowledge. Decision Analysis (DA) approaches translate available uncertain information on costs, benefits and risks involved in decisions into actionable management recommendations. We illustrate the use of DA procedures to inform decisions on disease management strategies in ornamental plant production. We worked with heather growers and other stakeholders in North Rhine-Westphalia, Germany, to model the impacts of changing disease management practices and to generate comprehensive forecasts of net returns. Through sensitivity analysis and Value of Information assessment we identified critical uncertainties regarding the feasibility of improved practices. Partial Farm Budgets for decision options ranged from a loss of more than 200,000 € to a gain of nearly 70,000 € per hectare and year. Findings suggest that reducing pesticide applications without additional monitoring may substantially increase production risks (chance of loss of 76%) and that intensified plant monitoring is likely to increase net benefits (chance of gain of 68%) by allowing earlier detection and more focused fungicide application. Our Decision Analysis approach facilitated ex-ante evaluation of innovative management strategies in heather production, and it holds promise for similar evaluations in other agricultural production systems.
