From model results to implementation in practice: the case of Mohair - - PDF document

From model results to implementation in practice: the case of Mohair Value Chain Sustainability Project

Lize Duminy1, Josephine Kaviti Musango2, Benjamin Batinge2, Camila Gillman3, Ian Taverner, Martin D Viljoen3, Suzanne Smit2

* Corresponding Author, Email: lize_duminy@hotmail.com 1Sustainability Research Area, Business School, Bern University of Applied Sciences, Switzerland 2School of Public Leadership, Urban Modelling and Metabolism Assessment (uMAMA), Centre for Complex Systems in Transition (CST), Stellenbosch University, South Africa 3South Africa Mohair Cluster, South Africa.

Abstract

The System dynamics modelling process provides a useful framework to capture the structure and decision-making process of a project. However, limited studies discuss how the model outcomes have resulted in impact or implementation. This paper’s objective is to demonstrate an exploratory stakeholder engagement process of moving from system dynamics model results, to implementation in practice, when the stakeholders’ priority is implementation, which entails, action planning, taking action steps and evaluating outcomes. The paper argues that system dynamics projects with implementation focus should be framed as an organisational intervention tool. A Cluster Intervention Framework – referring to the whole cluster of organisations involved in implementation was therefore developed, which is compatible to the Mohair Value Chain project implementation

• priority. As a way forward, system dynamicists need to provide empirical evidence of how model
• utcomes have resulted in implementation or impact, in order to establish a standardised modelling

framework for implementation.

Introduction

The system dynamics approach is versatile enough to be applied in a variety of fields. Conventional fields include research, business and public administration (Brent, et al., 2017) – in topics including everything from agriculture to motivation. Depending on the type of project and stakeholders involved, project success and desired output can be defined in a plethora of ways by each different stakeholder type. Assuming a system dynamicist produces a high quality, and validated model, other expert modellers, as well as academic reviewers of journals would usually receive it well. A positive reception is seldom contingent on whether the real problem was captured or whether the model resulted in any implemented change. Instead of focussing on academic dissemination of the system dynamics model and structural insights gained, this objective is to demonstrate an exploratory stakeholder engagement process of moving from system dynamics model results to implementation in practice. The paper describes a case highlighting that the technical quality of a system dynamics 1

model is not the only criteria defining the success of a system dynamics simulation project. The paper argues that literature on applied system dynamics should not exclusively focus on the structure of models developed but also incorporate how model results or ensuing insights should be implemented in practice.

Mohair Value Chain Sustainability Project Description

The project described is the first phase of a system dynamics simulation project with the objective of promoting global competitiveness and overall sustainability of the mohair value chain in South

• Africa. The Mohair Primary Production Echelon Model (MoPPEM) depicts the primary production

process of Angora goats with the output being mohair, carcass and skin. The South African Mohair Cluster commissioned the initiative after the Minister of Trade and Industry, Dr Rob Davis, named the Mohair Cluster among the clothing and textile sectors that would be promoted to enhance

• competitiveness. The South African Mohair Cluster brings together mohair industry stakeholders

with the objective of identifying opportunities and coordinating efforts to ensure competitiveness whilst promoting all three pillars of sustainability. The unique academic contribution of MoPPEM is that it is the first simulation study to examine mohair production in South Africa. System dynamics was deemed a compatible modelling tool since the mohair industry of South Africa exhibits boom-and-bust behaviours typical of commodity

• markets. There is also strong evidence that the behaviour of primary Angora goats and Mohair

production seem to be driven by feedback. Though widely used to model supply chains, there are limited applications of system dynamics in value chains (Pagani & Fine, 2008; Barbosa & Azevedo, 2017). The client, South African Mohair Cluster, approached the modellers regarding the potential to utilise system dynamics as a tool to help develop a shared understanding to engender coordinated strategic efforts between various industry stakeholders who previously did not see any value in

• cooperation. Once the project was launched, system dynamics was promoted to the mohair industry

stakeholders as a tool to visualise the major components of the mohair value chain and how each component is related with the goal of identifying leverage points for the benefit of the entire

• industry. A common concern for stakeholders was that other members of the mohair value chain

would use the model building process discussions against them in future. The most effective way of convincing the stakeholders with these concerns to participate was to utilise system dynamics as a ‘boundary object’ (Black, 2013). The participants were informed that the system dynamics approach

2

is based on the principle that structure drives behaviour, and that no individual actor or event would be discussed.

Insights from modelling and stakeholder processes

The Moxnes (2009) modelling process, Problem identification – Hypothesis – Analysis – Policy design – Implementation (P’HAPI), was used throughout model development. Multiple iterations of the Problem Identification, Hypothesis, Analysis and Policy phases were cycled through, in various combinations, throughout the project. Even though there was great excitement around the Implementation concept during stakeholder engagement, Implementation in the form of

• rganisational intervention was avoided in the final project report. The final report ultimately

contained relevant background on the mohair industry, argumentation as to why system dynamics is an appropriate tool for modelling the complexity within the mohair value chain, a description of the structure of MoPPEM, MoPPEM results, as well as scenario analysis. The Implementation phase of the P’HAPI methodology was not undertaken for two reasons. Firstly, the MoPPEM was only phase one of a larger project and implementation would only be suitable in later phases, and secondly the mohair industry is very hierarchical and politically loaded. This implied that interference from the project should be avoided until more information and consensus

• n the problem is gained to avoid refutation. Though the second reason was rational and valid, a

greater emphasis on practical, politically neutral guidelines could have given the modellers an advantage later. Both the modellers and stakeholders had an inherent bias regarding the nature of

• Implementations. Both parties expected Implementations to take on physical forms, for example, to

cultivate a local market for Angora goat meat among the South African middle class. However, since the beginning of the project it was acknowledged that substantial value-chain-wide organisational changes would be the most effective in achieving increased competitiveness and sustainability. The final report was well received by the Mohair Cluster of South Africa – with both the cluster and the modelling team impressed with the project phase I outcomes. However, some months later, before the second phase of the project was meant to kick off, the client realised the need for some, albeit generic, implementation recommendations stemming from MoPPEM to release the next phase of government funding. The modellers agreed greater dissemination and practical implementation recommendations could be achieved and the commencement of the next phase of the project was delayed to achieve the remaining requirements of phase one. The newly 3

communicated need translated into system dynamics serving as intervention architecture to increase organisational impact. The modelling teams’ behaviour during initial project development is intuitive since system dynamics modelling readily captures the decision process, not the implementation of decisions (Größler, 2007). However, implementation is important enough to be addressed in most system dynamics textbooks including Richardson and Pugh (1981), Roberts et al. (1983), Maani and Cavana (2000) as well as Sterman (2000). The study ultimately agrees with Coyle (1977) that implementation should be addressed throughout the project. A practical way of ensuring implementation is addressed throughout is to initiate the start of any new iteration of the modelling process with an Implementation phase, as recommended by Sterman (2000). Considering the implementation requirements in later stages of the project, the modellers conclude that the traditional system dynamics development frameworks (e.g. Randers, 1980; Richardson & Pugh III, 1981; Roberts, et al., 1983; Wolstenholme, 1990; Sterman, 2000; Moxnes, 2009) should be supplemented with elements of process consultation to frame system dynamics as an organisational intervention tool as the project matures (Rouwette & Vennix, 2006; Snabe & Größler, 2006). The modellers therefore adapted Schein’s (1969; 1999) organisational intervention framework: Cycle I entails finding a solution while Cycle II entails implementing the identified solution. The three steps

• f Cycle I - Problem Formulation, Producing Proposals for Solutions and Forecasting Consequences &

Testing Proposals – is compatible with the system dynamics modelling process already undertaken. In contrast, the three steps of Cycle II - Action Planning, Taking Action Steps and Evaluating Outcomes - is a disaggregated, hands on approach to implementation compatible with the needs of the project later on. Schein’s (1969; 1999) organisational intervention framework is therefore adapted to a custom Cluster Intervention Framework – referring to the whole cluster of organisations involved in

• implementation. The Cluster Intervention Framework is shown in Figure 1.

4

Figure 1: Cluster Intervention Framework

Cycle I represents the traditional system dynamics modelling framework used, in this case P’HAPI, without the Implementation step. Problem Identification, Hypothesis, Analysis and Policy Design and is shown as the inside layer of the framework in Figure 1. Cycle II remains Action Planning, Taking Action Steps and Evaluating Outcomes and represents the outer layer of the framework in Figure 1. In accordance with Schein (1969; 1999), Cycle I centres around diagnosing a problem and identifying a solution while Cycle II entails solution implementation. The outer layer will thus be the Cycle II of the Mohair Value Chain Project, whose focus will be entirely on Implemantation. The project will thefore continue with the preliminary iteration of Cycle II, with the output being generic implementation recommendations and discussions, before commencing with the second phase of model development, where MoPPEM is expanded to other echelons of the value chain. As a way forward, system dynamicists need to provide empirical evidence of how model outcomes have resulted in implementation or impact, in order to establish a standardised modelling framework for implementation. 5

References Barbosa, C. & Azevedo, A., 2017. Hybrid simulation for complex manufacturing value-chain

• environments. Procedia Manufacturing, Volume 11, pp. 1404-1412.

Black, L. J., 2013. When visuals are boundary objects in system dynamics work. System Dynamics Review, 29(2), pp. 70-86. Brent, A. C. et al., 2017. Utilization of System Dynamics in Southern Africa: A Systematic Review}. Systems Research and Behavioral Science, 34(6), pp. 657-670. Coyle, R. G., 1977. Management system dynamics. Chichester: Wiley. Größler, A., 2007. System dynamics projects that failed to make an impact. System Dynamics Review, 23(4), pp. 437-452. Maani, K. E. & Cavana, R. Y., 2000. Systems thinking and modelling: Understanding change and

• complexity. Auckland: Pearson.

Moxnes, E., 2009. Presidential address: Diffusion of system dynamics. Albuquerque, New Mexico, USA, The 27th International Conference of the System Dynamics Society. Pagani, M. & Fine, C. H., 2008. Value network dynamics in 3G-4G wireless communications: A systems thinking approach to strategic value assessment. Journal of Business Research, 61(11), pp. 1102-1112. Randers, J., 1980. Guidelines for model conceptualization. In: J. Randers, ed. Elements of the system dynamics method. Cambridge: MIT Press. Richardson, G. P. & Pugh III, A. I., 1981. Introduction to system dynamics modeling with DYNAMO. 1st ed. Cambridge: Productivity Press Inc.. Roberts, N. et al., 1983. Introduction to computer simulation: the system dynamics approach. Reading, Massachusetts: Addison-Wesley Publishing Company. Rouwette, E. A. & Vennix, J. A., 2006. System dynamics and organizational interventions. Systems Research and Behavioral Science, 23(4), pp. 451-466. Schein, E. H., 1969. Process consultation: Its role in organizational development. Reading, MA: Addison-Wesley. Schein, E. H., 1999. Process consultation revisited: building the helping relationship. Reading, MA: Addison-Wesley. Snabe, B. & Größler, A., 2006. System dynamics modelling for strategy implementation—case study and issues. Systems Research and Behavioral Science, 23(4), pp. 467-481. Sterman, J. D., 2000. Business dynamics: systems thinking and modeling for a complex world. Boston, MA: Irwin/McGraw-Hill. Wolstenholme, E. F., 1990. System enquiry: a system dynamics approach. Chichester: John Wiley & Sons, Inc.. 6