Building And Sustaining Resilience In Complex Organizations

110Moglia, Perez & BurnExample of social complexity in decision making: Water useand varying preferences: Preferences for what type of water to use fordrinking, cooking, cleaning, washing, etc, varies in the community. Forinstance, some community members, such as in the case study Tarawa,prefer the salty taste of the groundwater for their coffee, but this is nota dominating preference for the entire community. Heuristic modelsare able to capture the diversity of customer preferences in a range ofsituations.Water conservation and social capital: the capacity for water conservationis linked to social capital of the community, i.e. its willingness to cooperatefor the common good. See Rixon et al. (2006) for description of a modelrelating social capital and the potential for water conservation.Example of a non-linear feature: Pollution can render groundwatersources unfit for drinking (at least without significant and costlytreatment) for considerable time frames, and such pollution may insome cases happen in single instances.Examples of uncertainty: Drought and water scarcity: the mostobvious element of uncertainty relates to the coming and going ofdroughts that affect rainwater, groundwater and surface water resources.This can trigger water scarcity, which in turn has flow on effects such aswater allocation issues, and the spatial re-distribution of water resources(typically from rural to urban areas). See Ohlsson (2000) for a descriptionof the social conflict situations that arise from water scarcity.Example of social complexity (social acceptance): Many projectsin Tarawa seem to have failed because of lack of cooperation fromthe community. This seems to be due to a lack of consideration of theimportance of education and community engagement, as well as a lackof consideration of community needs in the planning process. Thisthen manifests itself as a risk of failure for any project due to unforeseencommunity behaviour. See Syme and Nancarrow (2006) for a descriptionof psychological considerations in social acceptance of reclaimed waterin an Australian circumstance.Examples of diverse socio-technical situations: The communityhas a range of socio-technical situations: In certain areas, it is possible touse local groundwater, while other areas have insufficient groundwaterlenses. Similarly, in some areas the community can use communalrainwater tanks, which requires an element of organization and socialcapital. In other areas, this would not be feasible.Table 1

How Complex is Water Service Delivery in a Small Atoll Town?111These requirements and constraints currently mean that decisions of significant complexity and uncertainty have to be made by asmall number of individuals. The way that individuals tend to makedecisions in a complex environment is to develop mental models, orhypotheses, from which to draw inferences and predictions of possibleoutcomes. In a system such as an urban water system, mental modelsare localized and depend strongly on the experience of the individual.Additionally, in an environment with inadequate information, complexity and uncertainty, the mental models based on traditional rationality are frequently burdened by incorrect assumptions and falsedeductions (Batten, 2000).Human resources, and strategic thinkingFor water service delivery in a small atoll town, the required knowledgebase for making many decisions consists of an understanding of:1. Water resources: balances, flows and sustainable extractionrates;2. Organizational capacity: skills, staff and culture;3. Financial issues: income, expenditure and customer base;4. Diversity of socio-technical settings: and related health, maintenance, financial and operational issues;5. Socio-political constraints: culture, preferences and politics;6. Environmental impacts: degradation, pollution and over-extraction;7. Infrastructure: degradation, capacity and network structure.Given this complexity, human resource issues represent anothersignificant challenge in small town water management; especiallywhen experienced staff are lured to more attractive jobs in bigger cities, or to jobs with better salaries or higher status.Leaving aside issues of staff retention, staffing constraints alsomean there tends to be a focus on the more operational issues thatappear on a day-to-day basis, rather than on strategic decision making. When water professionals must focus on day-to-day issues,there is no time to consider the important structural changes that arerequired to adapt to new conditions. Hence, strategic decision makingneeds to be accommodated in a manner that is not too burdensome

112Moglia, Perez & Burnon individual staff members, nor relies on individual staff membersfor continuity. To support this, there is a need for on-going capacitybuilding, knowledge management and provision of time for strategicdecision making.A conceptual management frameworkTo support strategic and adaptive decision making andimprove organizational resilience, a framework is neededfor capturing, validating and integrating a range of mentalmodels. To be sustainable, the framework should facilitate continuityof knowledge, and increase transparency of the decision makingprocess. The framework also needs to support adaptive management,and hence monitoring of the changing system states. This requiresthe elicitation of range of knowledge types, as well as a support forunderstanding system dynamics, uncertainty and inter-relations.To take these issues into consideration, a strategic and adaptiveframework is suggested as per Figure 1.This framework consists of four components:1. Adaptive management: continuous evaluation of policies andsystem states. This includes evaluation of wider types of information such as observations, stories and beliefs, which have theA daptiveMManaganagememententC apacityB uild ingS ystemRRepresentatioepresentationnF acilitatedD ialogu eFigure 1 Four-tiered framework for small town water management

How Complex is Water Service Delivery in a Small Atoll Town?113potential to provide rapid and rich information in situations wherehard data is limited2. Facilitated dialogue amongst practitioners and stakeholders: as a platform for common understanding ,which is requiredfor negotiations and social acceptance3. Capacity building: through the initiation of a permanent orsemi-permanent forum for integrated analysis and stakeholderand practitioners interaction. This would support the AdaptiveManagement framework, and rely on Facilitated Dialogue and anadequate System Representation4. Effective system representation: This is a key support block forthe other components. The requirements are that it needs to be anIntegrated analysis, taking complexity into account and supportingmulti-stakeholder interaction.It is noted that this framework is strongly aligned with GabrieleBammer’s call for a new specialization in Integration and Implementation Sciences (Bammer, 2005) that has its basis in ‘systems thinkingand complexity science’, ‘participatory methods’, and ‘knowledgemanagement’. Bammer compares this proposed specialization tostatistics and epidemiology because, just like these disciplines, Integration and Implementation Scientists will have mainly methodologicalexpertise rather than content expertise. It is also noted that this overallframework aims at improving the organizational resilience of the watermanagement organization.Various aspects of this conceptual management framework arediscussed in the following sections. The research approach beingundertaken in Tarawa to develop and promote such a framework ispresented in the final part of the paper.Adaptive managementBecause of the inherent uncertainty and dynamic nature of a watersystem, there is a need for Adaptive Management; which is aboutlearning from doing. This aims to increase the overall knowledgebase, in order to improve the quality of policies. Knowledge of failuresas well as success stories will then guide future policy making. Theframework for Adaptive Management is shown in Figure 2 (adaptedfrom Jakeman, et al., 2007).

114Moglia, Perez & BurnPolicy screeningDevelop systemunderstandingand modelingADAPTIVE MANAGEMENTCYCLEImplementingpolicies asexperimentsMonitoring andevaluating systemresponsesFigure 2 Adaptive Management Cycle(adapted from Jakeman, et al., 2007)As per Figure 2, the requirements for Adaptive Management are:1) having a system representation, and 2) the ability to monitor thesystem state and response to policies. In fact, Adaptive managementrequires continuous collection of information. Unfortunately, in acircumstance such as in Tarawa, where there is very limited capacityfor rigorous data collection, there is a need to consider many forms ofinformation and knowledge and diverse ways of knowing.Knowledge types and dialectical explorationTo overcome data problems and in order to allow for monitoring andevaluating system responses and system states, there needs to be anencompassing approach to knowledge, and a widening of the typesof knowledge considered. For instance, there are a number of typesof information that can be collected relatively easily, such as:1. Direct observations of conditions: for example, the observation of trees overhanging rain tanks, or level of vegetation onwater reserves;2. Records of health incidents: that reflect the health impacts ofwater service delivery in various socio-technical situations;3. Stories: of social interaction, socio-political situations, or seriesof events (for example relating to why the creation of additionalwater reserves and pumping galleries failed);

How Complex is Water Service Delivery in a Small Atoll Town?115S takeholders’S ystemR epresentatio n sIm pact onstakeholdersS takeholderactionsE volutio naryS ystem S tatesFigure 3 Feedback between system representation and system states4. Beliefs and mental models of stakeholders and participants:that can be drawn-out using various knowledge elicitation techniques.Direct observations and health records are relatively trust worthyin these circumstances (in terms of their ability to convey informationabout the true state), while stories, beliefs and mental models oftenpresent conflicting information. As such, there is no guarantee thatthey represent the truth, and this information needs to be treated withsome level of skepticism. On the other hand, exploration of disagreements (dialectical exploration) of such types of information oftenreveal beliefs and knowledge representations that actually impact onthe system state, as per Figure 3.Hence, when wider types of knowledge are considered, thereneeds to be stakeholder dialogue in order to increase the trust in theinformation used, while at the same time challenging and exploringstakeholder beliefs. Because Adaptive Management requires continuous evaluation of the system states, this dialogue should be withinsome kind of facilitated permanent forum.Cultural evaluationAdaptive management is also about exploring and evaluating deeprooted assumptions and cultural preferences. This is essentialfor a settlement’s organizational resilience, which is particularlyimportant at a time of cultural transition, such as is often thesituation in small towns in developing countries, where traditionalsubsistence farming and modern urban life coincide. In Tarawa, forexample, there appears to be a deep-rooted preference for coconuttrees and the consumption of pigs. Both these cultural preferencescan have detrimental effects on the water system. While thedecision regarding whether or not to keep such practices is up to the

116Moglia, Perez & Burncommunity, given the potentially wide reaching implications it canbe argued that this decision ought to be made in an informed andtransparent manner, rather than through passive acceptance.Facilitated dialogue and capacity buildingOne approach to overcoming staffing and knowledge managementissues is to set up a permanent forum allowing for on-going andfacilitated dialogue. Such a dialogue would have key advantages:1. Achieve a common system understanding through: Social learning; Dialectical exploration; Knowledge elicitation.2. Capacity building by: Interaction between individuals and the system representation; Providing a platform for knowledge continuity; Reducing reliance on single individuals.3. Strategic decision making supported by innovative explorationof a range of opportunities.Unfortunately, past experience shows there are hurdles in theimplementation of such forum. The first relates to the problem ofgetting stakeholders to participate. For example, previous attempts athaving a water forum failed in Tarawa. Sitting fees are often requiredin order to find participants, and even then participation may be lessthan efficient.A second hurdle relates to the complex socio-political settings, aswell as language issues, meaning that facilitation needs to be carriedout by an individual with a good knowledge of language and culture.This is a problem because of the commonly limited capacity of localNGOs to carry out such a task.Requirements for effective system representationTo support the other components of the conceptual framework,a system representation needs to be developed. This is in order toexplore and design policies before implementation, and therefore

How Complex is Water Service Delivery in a Small Atoll Town?117Demand management example: because water is often a commonpool resource, and especially in a situation where water demand ison par with water supply, there is a need for demand management. Inmarket economy societies, charging for water is typically used to reducewater demand, but this tends to require a relatively strong organization,including a customer database, a legal system to deal with non-payment,as well as the ability for customers to pay. In circumstances where manycustomers are not able to pay and where there is not a strong organization(such as in Tarawa, the field study), the situation is different. Thisrequires alternative methods for demand management be adopted suchas providing alternative sources of water, and communal solutions.Water safety example: Water safety depends on the complete path ofthe water from source to use (AusAID, 2005). In Tarawa, the safety ofthe water pumped from groundwater infiltration galleries depend on: Activities on groundwater reserves such as settlements, croppingand industrial activity - Whether there is 24 hour pressure in thepumping mains; The chlorination schedule; Water storage at household level and water resource level; How the water is finally used.Infrastructure deterioration example: Infrastructure deteriorationis often serious in Tarawa because of the limited ability to deal withinfrastructure failure. This relates to inability to source spare parts, as wellas a lack of funds and capacity for repair. For example, three desalinationplants were originally installed in Tarawa with a total capacity of 200 m3per day. A combination of factors, such as the rapid degradation due tothe highly corrosive environment, and the inability to source spare partsfrom the Chinese manufacturers, has led to all three desalination plantscurrently being inactive.Table 2planners need a system representation in order to evaluate the chancesof success, as well as the likely impacts of policies. As a secondaryadvantage, a system representation can serve an educational purposeand could therefore be useful in capacity building. A meaningfulsystem representation will allow decision makers to at least considera multitude of perspectives and impacts. Some examples are givenin Table 2 of the issues that must be taken into account within an

118Moglia, Perez & Burneffective system representation.Since there are so many complexities and challenges associated withwater resource management for small atoll towns, it is perhaps unrealistic to think that any single person can fully grasp the whole systemand surrounding issues. To support decision making it is thereforesuggested that a system representation needs to be developed that isshared and agreed upon by as wide a representation of stakeholdersas is feasible. The requirements of such a system representation arethat it needs to:1. Integrate a multitude of perspectives, i.e., it needs to supportIntegrated Analysis;2. Represent the dynamic behavior of the system, in terms of theAdaptive behavior, Social Complexity, Non-linear features andUncertainty;3. Represent a diversity of socio-technical situations (such as in adistributed model);4. Be useful in supporting Facilitated Dialogue.All these requirements point towards the development of an AgentBased Model (ABM) for system representation. In accordance withconcerns raised by Gintis (2006), it is however acknowledged thatAgent-Based Modelling provides an experimental framework withsome limitations, rather than being a theoretical model. For example,whilst an ABM has many advantages, there are issues relating to theability for replication (two different modelers may get two differentresults despite the same specification) and validation. As Richardson(2002) has pointed out, there are also issues of equifinality (i.e., therecan be multiple non-overlapping explanations of the same phenomena), and there is a need for weak (intra-perspective and quantitative)and strong (inter-perspective and qualitative) exploration.Whilst acknowledging these limitations, the researchers believethat these issues are adequately addressed through a process of iterative confrontation between model, reality and multiple stakeholderperspectives (see the Companion Modeling section below).

How Complex is Water Service Delivery in a Small Atoll Town?119Current researchResearch is currently being undertaken to investigate andpromote the implementation of the proposed strategic andadaptive management framework using Tarawa as a casestudy.Case study situationAs per Diamond’s risk framework (Diamond, 2005) mentionedpreviously, Tarawa is at high risk because of:1. Population increases and urban developments: which have ledto environmental degradation and loss of traditional livelihoods;as well as conflicts in relation to land-use, and land ownership.2. Insufficient response to environmental changes and damage: partly because of a transition to an urbanized and westernized style of environmental management, but in a situationof severe constraints; commonly because of a strong preference for centralisedand inflexible large scale technical solutions; sometimes related to the unrealistic urban customer expectations.3. Significant reliance on trade partners: such as Australia andNew Zealand.4. Climate change: with an increased frequency of El-Nino eventsand related droughts, as well as the long term threat of rising sealevels.The availability of groundwater underpins the viability of settlements in Tarawa. However, concerns surrounding the water reserveshave been extensive due to the limited availability of land, and rightsof traditional land owners. Therefore, social acceptance

In t mes of rap d change, such as currently caused by urban zat on, global zat on and cl mate change, the