Optimizing Production Decisions Using a Hybrid Simulation–Genetic Algorithm Approach

Mathematical programming has for a long time been recognized as a powerful tool. Despite its capacity for solving constrained optimization problems under uncertainty, some methodological obstacles have persisted over the years. The main problem is that the eventually complex results of an unbiased statistical analysis (multiple correlated stochastic variables with different distributions and nonadditive links between) cannot be adequately accounted for within minimization of total absolute deviation (MOTAD) or expected value‐variance (EV) models that rely on the algorithmic determination of the variability measure. In this paper, we develop a methodological hybrid consisting of Monte Carlo simulation and genetic algorithms: the Monte Carlo simulation facilitates the easy representation of diverse stochastic processes and correlation, and the genetic algorithm ensures that the optimization procedure remains applicable even in the case of complex stochastic information. This hybrid approach is applied to the production‐planning problem of a German crop farm. Variant calculations are used to account for the unknown risk attitude of the farmer. Model results demonstrate that optimized production programs and expected total gross margins are not only highly sensitive to the risk attitude, but also to the stochastic processes that are estimated (or assumed) for various activities. We furthermore find evidence that the hybrid approach is able to generate considerable improvement in farm‐program decisions and outperforms planning models that assume static distributions. La programmation mathématique est reconnue depuis longtemps comme étant un outil puissant. Malgré sa capacitéà résoudre des problèmes d'optimisation avec contraintes en situation d'incertitude, certains obstacles méthodologiques ont persisté au fil du temps. Le principal problème réside dans le fait que les résultats éventuellement complexes d'une analyse statistique non biaisée (plusieurs variables aléatoires corrélées avec différentes distributions et des liens non additifs entre elles) ne peuvent être adéquatement représentés dans les modèles MOTAD et E‐V qui dépendent de la détermination algorithmique de la mesure de la variabilité. Dans le présent article, nous avons élaboré une méthode hybride à partir d'une simulation de Monte Carlo et d'algorithmes génétiques: la simulation de Monte Carlo facilite la représentation de divers processus stochastiques et de diverses corrélations, et l'algorithme génétique assure que la procédure d'optimisation demeure applicable même dans le cas d'information stochastique complexe. Cette méthode hybride est appliquée au problème de planification de la production auqúel est confrontée une exploitation de cultures en Allemagne. Des calculs de variantes sont utilisés pour tenir compte de l'attitude inconnue du producteur quant au risque. Les résultats du modèle indiquent que les programmes de production optimisés et les marges brutes totales prévues ne sont pas uniquement sensibles à l'attitude face aux risques mais aussi aux processus stochastiques qui sont estimés (ou supposés) pour diverses activités. Nous avons également trouvé que la méthode hybride peut améliorer considérablement les décisions concernant les programmes agricoles et qu'elle est supérieure aux modèles de planification qui supposent des distributions statiques.     

A Linear Programming Algorithm for Determining Mean-Gini Efficient Farm Plans

A linear programming algorithm is presented for determination of farm plans that are efficient under the mean-Gini criterion. Use of the algorithm is illustrated by applying it to a standard problem from the literature. The set of efficient plans is compared with those based on mean-variance and MOTAD analysis of the same problem.     

Farm Capital Investment and Deviations from the Optimal Path

We identify farms’ optimal investment path in capital assets and compare it with their actual investment to assess the direction and extent of deviation from the optimal investment. A probit model is further used to investigate the determinants of the probability that a farmer over‐ or under‐invests in capital assets. We use a panel dataset of Dutch dairy farms over the period 2003–2013, and find that most farms under‐invest in capital assets during the study period. Although the number of farms that had over‐invested in capital assets is relatively small, these farms account for the biggest share of total investment in capital assets. The probit results show that liquidity, agricultural support payments, age, land tenure and standard output size are important variables explaining the likelihood of over‐and under‐investment.     

Valuing Pre-Commercial Genetic Resources: A Maximum Entropy Approach

Genetic improvement has been a major contributor to agricultural productivity in the United States, but many questions about the economics of crop breeding, such as the value of pre-commercial germplasm, remain unanswered. This study estimates the marginal value of poorly characterized materials contained in the U.S. national germplasm system. Within the search theoretic framework, we apply a maximum entropy method to estimate the probability and the expected level of improvement in pest susceptibility relative to its best previously observed level. The results indicate that the lower-bound estimate of benefit is significantly higher than the upper-bound cost of conserving an accession.     

On‐Farm Conservation of Rice Biodiversity in Nepal: A Simultaneous Estimation Approach

This paper presents an empirical case study about farmer management of rice genetic resources in two communities of Nepal, drawing on interdisciplinary, participatory research that involved farmers, rice geneticists and social scientists. The decision‐making process of farm households is modelled and estimated in order to provide information for the design of community‐based conservation programmes. A bivariate model with sample selection examines the simultaneous process of whether farmers decide to plant landraces or modern varieties, and whether the landraces they choose to plant constitute the genetic diversity of interest for future crop improvement. Findings show that the two landrace choices are affected by different social and economic factors, but in certain cases the decision processes are interrelated. Policies to promote the conservation of local rice diversity will need to take both processes into account.     

Exploring Farm Investment Behaviour in Transition: The Case of Russian Agriculture

This paper analyses the investment behaviour of Russian farms during the period of economic stabilisation that followed Russia’s financial crisis of 1998, and is the first to apply the error‐correction investment model to describe farms’ investment behaviour in the transitional context. Additionally, the paper employs the error‐correction and the adjustment‐cost model to test for differences in the investment behaviour between various farm categories. The results show that in general Russian farms exhibited an error‐correcting behaviour in the period under investigation. From 1999 to 2005 the output–capital gap was closed by an average rate of 10% per year. Estimates of the adjustment‐cost model show that Russian farm investments are very sensitive to the sales–capital ratio, suggesting that Russian farms exhibit increasing returns to scale and positive expectations about future revenues. Yet, such farm characteristics as ownership structure, access to input markets and also regional specifics were found to be decisive for farm investment not only in the short but long term too. Finally, the results show that the adjustment‐cost model is adequate for the evaluation of differences in short‐term investment behaviour, whereas it is noticeably less powerful for investigating differences in the farms’ long‐term investment behaviour.     

国有林场职工租赁经营果园的探讨

本文着重介绍平和天马国有林场立足山地资源 ,引导职工种植果树 ,发展经济的历程。探讨国有林场职工租赁经营果园的路子、存在的问题及对策     

Does Capital Market Structure Affect Farm Investment? A Comparison using French and British Farm–Level Panel Data

This article considers whether differences in the structure of agriculture credit markets in France and the United Kingdom alters the investment sensitivity to financial variables particularly cash flow. Using two panel datasets of French and British farms, three approaches are used to test the sensitivity of investment to internal finance, an inventory investment model, a fundamental q –model, and Euler equations for machinery investment. The results suggest that the contrasting capital markets structures do induce differences in overall investment sensitivity to cash flow and its pattern across both farms with varying levels of collateral and between inventory and machinery investment.     

Capital Investment under Alternative Marketing Scenarios in the Hog Industry: A Real Option Approach

This paper uses a real option approach to analyze the impact of alternative marketing contracts on the decision to invest in a cooperatively owned hog facility. For the numerical analysis of the impact, this paper uses a simulation method that incorporates early exercise, multiple‐state variables, multi‐choice decisions and temporal optimality. The results show that the option values that stem from the value of waiting to invest and choosing between alternative marketing methods amounts to 20–36% of the initial investment. Further, having an option to choose an alternative marketing method with different risk structure does add to the value of waiting to invest. Having an option to enter a 15‐year marketing contract increases the value of waiting by as much as $117,097 for the pork production example in this paper. Finally, the value of the option to wait is unilaterally lower under a risk‐reducing contract scenario than under a spot market alternative. This could explain the explosion in hog production facility investment during the 1990s when prevalence of contract production increased. Les auteurs ont recouru à l'approche du choix véritable pour analyser l'impact d'autres solutions de mise en marché sur la capacité d'investir ou pas dans un élevage de porcs exploitéà la manière d'une coopérative. Aux fins de l'analyse, on s'est servi d'une méthode de simulation intégrant une brève campagne, de multiples variables d'État, des décisions à choix multiple et des conditions optimales dans le temps. Les résultats indiquent que l'existence d'un choix réel permettant à l'exploitant d'attendre avant d'investir et de sélectionner entre plusieurs méthodes de mise en marché représente 20 à 36 pour cent de l'investissement initial. Par ailleurs, le fait d'avoir accès à une autre méthode de mise en marché, à structure de risque différente, ne donne pas plus de valeur à la capacité d'attendre avant de procéder aux investissements. La possibilité de signer une entente de commercialisation de quinze ans accroît la valeur de l'attente de jusqu'à 117,097 $ pour le type d'élevage porcin retenu comme exemple. Enfin, l'existence d'une entente de commercialisation atténuant les risques réduit unilatéralement la valeur de l'option « attente », comparativement à ce qui se produit quand l'éleveur n'a d'autre choix qu'écouler ses bêtes sur le marché au comptant. Ces résultats pourraient expliquer l'explosion des investissements observée dans le secteur du porc au cours des années 90, où la production sous contrat avait sensiblement augmenté.     

A Real Options Approach for the Investment Decisions of a Farm‐Based Anaerobic Digester

The profitability of anaerobic digesters (ADs) for Ontario dairy farmers are examined using real options under current and proposed government pricing policies and investment uncertainty. In the case of a renewable energy initiative such as an AD with large sunk costs and volatile returns, the value of deferring investment may be significant enough to offset the returns suggested by the net present value (NPV) approach. For a 150 cow herd, net revenues should be approximately $1.1 million before the AD is installed using the real options approach as compared to $0.5 million with the NPV approach. An AD is close to generating a positive NPV for a 600 cow herd if for either a 1% increase in the electrical price or decrease in the cost. However, farmers need not invest today and there is a value to delaying this decision from potential improvements in the technology that increase the efficiency and/or decrease operating costs of the AD. The real options analysis indicates that this option to delay investment has a value of approximately $300,000 for a typical Ontario dairy farm. Thus, either significant grant funding or higher feed‐in‐tariff rates are required to induce the increased adoption of AD technology in Ontario today even for the largest of dairy farms. Considering the probability of government support potentially ending, increases the value of investing today but a significant option value to defer still exists.     

福建省国有林场营林投资经济效益评价研究

运用净现值法和内部收益率法,用福建省国有林场杉木生产过程中发生所有费用和木材生产过程中发生费用的资料,研究福建省国有林场营林投资经济效益。结果表明:国有林场林业投资有一定的经济效益,但经济效益还是比较低;以6%的资本利率,杉木投资的净现值为178.53元/hm2。林业税费政策调整以后,林业投资收益主要取决于造林抚育的支出水平。     

Capacity Depreciation, Implicit Rental Price, and Investment Demand for Farm Tractors in Canada

Among the different factors pertaining to the demand for durable inputs in agriculture, the way the input depreciates over time is an important issue. This study investigates four alternative capacity depreciation patterns for farm tractors. While published estimates from government agencies implicitly assume convex capacity depreciation patterns, the results of this study demonstrate that the use of agricultural engineering data which depict a concave rather than convex pattern better reflect farmers' investment decisions. An implicit rental price measure is also proposed to account for the fact that the purchase price of a durable input is not an appropriate measure of the implicit rental cost of the asset.
Parmi les facteurs déterminants de la demande pour les biens durables en agriculture, la rapidité avec laquelle I'intrant se déprécie est certainement importante lors des décisions d'investissement. La présente etude analyse quatre différents scénarios de dépréciation pour les tracteurs agricoles au Canada. Bien que les données publiées par les agences gouver-nementales supposent implicitement que les biens durables se déprécient à un taux annuel constant, les résultats de cette etude montrent que I'utilisation de données basées sur la valeur productive de I'intrant telle que calculée sur une base expkrimentale donne de meil-leurs résultats statistiques dans I'estimation de la demande pour les tracteurs agricoles. De plus, une mesure du coût implicite de I'intrant est également proposée pour tenir compte du fait que le prix d'achat n'est pas une mesure appropriée dans I'analyse de la demande pour un bien durable.     

Effects of Farm Size on Technical Efficiency in China's Broiler Sector: A Stochastic Meta‐Frontier Approach

This paper employs the stochastic meta‐frontier approach to measure technical efficiency and to investigate the effects of farm size on the technical efficiency in China's broiler sector. Empirical results show a positive association between farm size and technical efficiency in China's broiler sector. The medium and large farm sizes exhibit increases of 0.058 and 0.160, respectively, in technical efficiency scores, relative to small farms, which have a mean technical efficiency score of 0.722. Results indicate that there are significant differences in technical efficiency across regions. Technical efficiency in the southern region, which is dominated by yellow‐feathered broilers, is significantly lower than that in the northern region where white‐feathered broilers are the dominant species. Also, the technical efficiency scores estimated from the meta‐frontier model vary substantially across farm sizes in the southern region. Increased farm size improves the technical efficiency for yellow‐feathered broiler production. Dans la présente étude, nous avons utilisé l'analyse métafrontière stochastique pour mesurer l'efficience technique et pour déterminer les effets de la taille de la ferme sur l'efficience technique du secteur du poulet à griller en Chine. Les résultats empiriques indiquent qu'il existe un lien positif entre la taille de la ferme et l'efficience technique. Les fermes de moyenne et grande taille affichent une augmentation du pointage d'efficience technique de 0,058 et de 0,160 respectivement, comparativement aux fermes de petite taille dont le pointage d'efficience technique moyen est de 0,722. Les résultats indiquent que l'efficience technique varie considérablement d'une région à l'autre. L'efficience technique dans la région du Sud, où l’élevage du poulet à griller blanc domine, est significativement plus faible que celle de la région du Nord, où l’élevage du poulet à griller jaune domine. De plus, les pointages d'efficience technique estimés à partir du modèle de métafrontière varient considérablement dans la région du Sud. L'augmentation de la taille de la ferme améliore l'efficience technique de la production du poulet à griller jaune.     

Farm Input, Farm Output and Retail Food Prices: A Cointegration Analysis

This paper investigates the adjustment mechanism between farm input prices, farm output prices and food retail prices in Canada. Johansen's maximum likelihood approach is used in addition to the Engle-Granger approach to test for cointegration. Contrary to the common assumption that farm output prices are more flexible than farm input prices, it is found that farm output prices, though cointegrated, are weakly exogenous in the sense that they do not respond in a systematic manner to disequilibrium in farm input prices and retail food prices. Evidence is found to support “cost push” and “demand pull” theories but, since food retail prices carry a heavier weight in the cointegration relations, it can be concluded that shocks manifesting themselves (first) at the retail level do not persist as long. Cet ouvrage se concentre sur le mécanisme d'ajustement entre les prix des intrants à la ferme, les prix payés aux producteurs agricoles et les prix de l'alimentation au détail au Canada. Les techniques développées par Johansen et par Engle et Granger sont utilisées pour vérifier des hypothèses reliées à la cointegration. Contrairement à la présomption que les prix payés aux agriculteurs sont plus flexibles que les prix de leurs intrants, on a découvert que les prix payés aux agriculteurs, bien que cointégrés, sont exogènes au moindre degré dans le sens qu'ils ne s'ajustent pas en fonction du déséquilibre dans les prix des intrants et les prix de l'alimentation au detail. La théorie voulant que les prix montent à cause des hausses dans les prix des intrants est vérifiée de même que la théorie alternative voulant que les ajustements des prix se fassent du détail aux intrants. Fluctuations occasionées par des changements dans les prix de l'alimentation au détail sont dissipées plus rapidement que les ajustements causés par des variations dans les prix des intrants.     

Estimating Farm Productivity Differentials using Panel Data: The Hausman‐Taylor Approach

This paper investigates the relation between unobserved farm productivity and other production factors in a system of netput equations for specialised pig breeding farms in the Netherlands. In order to estimate the system, a Hausman‐Taylor panel data estimator is developed for a system of equations with unbalanced panel data. Tests on the correlation between model variables and farm‐effects are performed, yielding an insight into the sources of differences in total factor productivity and its components (e.g. managerial ability and scale economies). Results indicate that specialised pig breeding farms that are characterised by high total factor productivity have more buildings and machinery than farms with low total factor productivity.