ENBIS Spring Meeting 2018

4 – 6 June 2018; Florence, Italy Abstract submission: 17 November 2017 – 20 April 2018

My abstracts


The following abstracts have been accepted for this event:

  • Perspective of sustainable climate service for small scale farmers in Africa

    Authors: Moammar Dayoub (Turku University), Erkki Sutinen (Turku University), Mikko Apiola (Turku University), Jaakko Helminen (Turku University), Ville Myllynpää (Turku University)
    Primary area of focus / application: Business
    Keywords: sustainable climate service,, small scale farmers,, co-design,, Africa
    Submitted at 17-Apr-2018 11:44 by Moammar Dayoub
    Climate variability is one of the most complex problems the world faces. It especially affects the most vulnerable people in the Global South. In the Sustainable Development Goal (SDG) number two, countries agreed to end hunger, achieve food security and improved nutrition, and promote sustainable agriculture by 2030. The aim of the research study is to offer climate information and Information and Computer Technology, ICT, based solutions for small scale farmers and other interest groups such as government officials and decision makers to adapt to the effects of climate change in food production.
    The main focus of the project is using a co-design based development of a mobile climate service application aimed at the local farmers in Tanzania. In the second phase of the project, the results will scaled up also in other areas. The application built the functions, which shares climate knowledge (based on the climate and weather data) and provides knowledge about new farming techniques, best practices and adaptation measures to local farmers. It will also help farmers to use this knowledge to choose crops and seeds, growing methods, as well as sowing and harvesting times, which are better suited for the changing climate. It can also include additional features, such as insurance against drought, market platform for selling extra harvest and even funding tool to allow farmers to invest into new growing techniques. This process will eventually help farmers improve their income generation, through improved yields and higher cash crops.
    At the core of the whole project and one of its key added values is co-design; it implies the continuous exchange of ideas, the co-development of the climate service in a coordinated way and a co-learning process as the stakeholders will create an open platform in a truly participatory method with the local stakeholders. One important goal is to enhance the potential of small scale farming and its capacity to offer improved livelihood to a growing number of people. Making farming interesting for new generations by introducing new innovative methods will make small scale farming more attractive to young people, who now flee to the cities in hope of better livelihood and too often end up in the streets or migrate. It will also ensure that small scale food production has a future and ensure food security in the most vulnerable areas.
    Sustainable business research will be used to help developing both a sustainable revenue and a business models for the application and providing farmers´ additional income sources, through improved agriculture practices. The aim is to develop a sustainable business model, which will allow the project to sustain itself and continue to operate without external funding, after the initial project development funding has ended. This is important, since nowadays, unfortunately, many promising projects fail after the funding period ends.
    We need to support various business opportunities, both for local companies and organisations, as well as for international companies, by providing research knowledge about climate knowledge services in the agriculture sector.
  • Designing and conducting discrete choice experiments with the R-package idefix

    Authors: Martina Vandebroek (KU Leuven), Frits Traets (KU Leuven)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Business
    Keywords: Discrete Choice Experiments, Design of experiments, R-package, idefix
    Submitted at 18-Apr-2018 10:36 by Martina Vandebroek
    5-Jun-2018 12:20 Designing and conducting discrete choice experiments with the R-package idefix
    Discrete choice experiments are widely used in a broad area of research fields to capture the preference structure of respondents. The design of such experiments will determine to a large extent the accuracy with which the preference parameters can be estimated. This presentation presents a new R-package, called idefix, which enables users to generate optimal designs for discrete choice experiments based on the multinomial logit model. In addition, the package provides the necessary tools to set up online surveys with the possibility of making use of the individual adaptive sequential Bayesian design approach for estimating the mixed logit model. After data collection the package can be used to transform the data into the necessary format in order to use existing estimation software in R.
  • Strategies for employing DOE to develop processes with variation in customer use

    Authors: Jacqueline Asscher (Kinneret College)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Consulting
    Keywords: DOE, strategy, factors, levels, customer use
    Submitted at 18-Apr-2018 10:44 by Jacqueline Asscher
    4-Jun-2018 17:05 Strategies for employing DOE to develop processes with variation in customer use
    There are myriad sources of variation in our experiments, including raw materials, process conditions and parameters, adjustments made by operators, measurement systems and customer use. We focus here on the latter, considering three different scenarios. For each scenario, we describe the problem and ask: How should we choose factors, factor levels and response variables in our experiments? What information do we need in order to make these choices? How do we obtain this information?
    Common to all three scenarios is the nature of the product. It is used by the customer with either a variety of their own products, or under varying conditions, or both. For example, our printer is used by the customer to print a variety of fabrics with a variety of inks; our packaging system is used by the customer to pack different types of produce at varying conditions of temperature and humidity, and then stored under different conditions; our plastic agricultural product is used in a variety of fields or orchards, etc. Note that conditions of use include both random and fixed effects, for example, in the case of the ink or the produce, there are both types and batches of ink and produce.
    Scenario 1: The product/process has been developed. We aim to characterize the conditions of use e.g. define “our product works at temperatures below 40C”, “our system can be used to package the following types of produce…”
    Scenario 2: A completely new product/process is being developed. We aim to both enable it to work with a wide range of conditions of use and to characterize the conditions of use.
    Scenario 3: We have a range of existing products/processes, with available field data from beta testing, complaints, and service records. A new product/process, similar to existing ones, is being developed. We aim to both enable it to work with a wide range of conditions of use and to characterize the conditions of use.
    This talk includes an opportunity for conference participants to discuss the issues raised.
  • An Integrated Approach For Measuring Matching And Strength In Green Brand Associations Through Text-Mining Techniques And Log-Linear Models

    Authors: Silvia Ranfagni (Department of Economics and Management, University of Florence), Nedka Dechkova Nikiforova (Department of Statistics, Computer Science, Applications “G. Parenti”, University of Florence)
    Primary area of focus / application: Business
    Keywords: Green Brand, Green Brand Associations, Sustainability, Text-Mining, Hybrid Log Linear Models
    Submitted at 19-Apr-2018 14:15 by Nedka Dechkova Nikiforova
    6-Jun-2018 09:55 An Integrated Approach For Measuring Matching And Strength In Green Brand Associations Through Text-Mining Techniques And Log-Linear Models
    Nowadays, an increasing number of companies adopt green brand strategies by prioritizing sustainability in their agenda. To this end, an important issue to consider is how to efficiently measure the green brand associations that compose the brand image as perceptual nodes that consumers associate with the green brand and hold in their memory. With this in mind, in this talk we propose an innovative approach to estimate the alignment between company-defined and consumer-perceived green brand associations in virtual settings by integrating linguistic tools with log-linear models. More precisely, we propose an integrated approach in which textual data and Hybrid Log Linear models are combined in a whole, interconnected procedure in order to measure the alignment between the voice of the company and the voice of the customers in terms of green brand associations. In order to develop it, we have investigated online communities related to a green fashion brand. We have identified the common co-occurrences of both company and consumers and related to green brand associations through text-mining techniques. Following, these common co-occurrences have been analyzed through Hybrid Log Linear models that result particularly useful for studying complex structures of associations formed by co-occurrence counts in a contingency table (Goodman, 1965; Bishop et al., 1975; Fingleton, 1984). Moreover, our proposed approach allows to obtain a unified framework for measuring: i) consumers vs. company brand association matching, and ii) the strength of consumers vs. company green brand associations, by also providing an innovative marketing tool for defining green brand strategies.

    1) Goodman L. A. (1965). On the Statistical Analysis of the Mobility Tables. American Journal of Sociology, 70(5): 564-584.
    2) Bishop Y. M., Fienberg S. E. and Hollande P. W. (1975). Discrete Multivariate Analysis: Theory and Practice. Cambridge: MIT Press.
    3) Fingleton B. (1984). Models of Category Counts. Cambridge: Cambridge University Press.
  • Bayesian sequential design approach for multi-objectives applications

    Authors: Matteo Borrotti (Energia Crescente S.r.l. and CNR-IMATI)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Sequential experimental design, Multi-objective optimization, Pareto optimality, Bayesian framework
    Submitted at 20-Apr-2018 10:45 by Matteo Borrotti
    5-Jun-2018 16:55 Bayesian sequential design approach for multi-objectives applications
    Nowadays, technologies and innovative tools allow to experiment complex problems characterized by an increasing number of variables and system responses. In this context, variables share non-linear relations and system responses conflict with each other. Optimize a solution with respect to a single system response can lead to unacceptable results. The process to find optimal solutions for a specific problem has never been so difficult as today. In this work, a Sequential Experimental Design (SED) based on Bayesian framework suitable for multi-objectives problems is proposed. The approach is tested on a set of simulated case studies and is compared with state-of-the-art techniques assess main advantages and disadvantages of the proposed solution.
  • Experimental designs suitable for cases with varying error variances

    Authors: Garima Priyadarshini (Imperial College London)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Modelling
    Keywords: Design of experiments, Heteroscedasticity, Sparse design matrix, Efficient designs
    Submitted at 20-Apr-2018 13:54 by Garima Priyadarshini
    For the usual linear model given as Y= Xβ + ε, the ordinary least squares (OLS) estimator is BLUE (best linear unbiased estimator) only if the assumptions of the linear regression are satisfied. Out of these, the assumption of homoscedasticity is observed to be violated frequently and certain solutions to deal with it are provided as well. These solutions, however, assume some knowledge about the form of the variance terms. As estimation of treatment effects in experimental designs is essentially OLS estimation, and any knowledge about the form of the observed variances is quite unlikely to be there at the time of designing an experiment for a process, it would be useful to obtain experimental designs specifically suited for the case of variable error variances.
    This work aims at defining a methodology to obtain suitable experimental designs for processes that exhibit random error variances. The methodology proposes designs with sparse associated design matrices, tailored to facilitate estimation of the treatment effects of interest. As presence of heteroscedasticity makes the treatment effect estimates non consistent, the utility of these designs lies in the fact that the estimate of the variance of treatment effect estimates are brought close to their OLS counterpart by means of utilizing this sparsity in the design matrix. The proposed designs are, thus, the most efficient for estimating the effects of interest when the error variances are not same.