Evolution of a Dynamic Theory of Collaboration Modeling ...

Evolution of a Dynamic Theory of Collaboration: Modeling Intergovernmental Use of Information Technology Prepared for the 2002 System Dynamics Research Conference, Palermo, Italy Presented at the Conference by: Dr. David F. Andersen Ignacio J. Martínez-Moyano Center for Technology in Government / Rockefeller College University at Albany July 2002 Authors Anthony Cresswell tcresswell@ctg.albany.edu Laura Black lblack@ctg.albany.edu Donna Canestraro dcanestraro@ctg.albany.edu Meghan Cook mcook@ctg.albany.edu Theresa Pardo tpardo@ctg.albany.edu Fiona Thompson fthompson@ctg.albany.edu Center for Technology in Government (CTG) University at Albany Albany, NY 12222 Luis Luna ll8287@albany.edu Ignacio Martinez-Moyano im7797@albany.edu David Andersen david.andersen@albany.edu George Richardson gpr@albany.edu Rockefeller College of Public Affairs and Policy University at Albany Albany, NY 12222 Contents • Where we’ve been: Project history and critical theoretical problems • Where we are: Approaching solutions • Where we might go: Toward a dynamic theory of collaborative IT Where We’ve Been: Project History and Some Problems with Representation and Theory Motivation • Interorganizational partnerships are widely recognized as a powerful strategy to improve public sector initiatives in order to significantly increase the quality of their services. • Information Technology plays a key role in this partnerships. Motivation • Researchers at the Center of Technology in Government have studied knowledge and information sharing in interorganizational networks in the Public Sector for years. • Their analyses have provided evidence of the existence of feedback processes influencing collaboration and knowledge sharing. • A dynamic theory of collaboration could be a powerful instrument to improve success in IT intensive projects. Motivation • The creation of better understanding and better models of interorganizational dynamics. Time Line Trust1 2 Modeling Sessions January 2001 KnowledgeCollaboration 1 Collaboration 2 Based Trust 1 Modeling Work March 2002 June 2002 First Model Structure Elicitation (Original image redrawn in Vensim) Pres sure to be acc ountable Welfare reform pressure Bob used Bob activity negative experienc e Feasible prototype component components growth Capacity to collaborate Pers onal prior experienc e Expectations Demonstrated results Opportunity to act Common understanding of what and how Use of SMART Role of IT tools corporate partner CTG involvement Responsibiliy for collaboration Willingness to collaborate Collaboration Trust Leadership Provider Engagement BHS and QA engagement Reflector Feedback Model Sectors Project Sector    Feasible Components Unfeasible Components Demonstrated Results Demonstrated results build engagement Demonstrated results build engagement State and BHS Collaboration   BHS engagement QA engagement Providers Sector     Committed Providers Uncommitted Providers Level of commitment of the committed Trust CTG Sector    CTG effort on collaboration CTG effort on trust building CTG effort on skill building Unsolved problems Unsolved problem generation Unsolved problems per component Average unsolved problems per component Projec t definition Perc eived progres s fraction Trust 1 Effec t of average problems on satis faction Satisfaction in demonstrated results Effec t of collaboration on Unresolved Components Feasible Components Progress rate Productivity People on project development Willingnes s to adjust workforce Effec t of Collaboration on Produc tivity Available State Effort Time to perceive potential Potential State Effort BHS and QA engagement Available people Perc eived Potential Collaboration Available providers effort Maximum effort per provider Indicated engagement Engaging Potential provider effort Indicated state engagement Effec t of CTG effort on c ollaboration Building Engagement of committed providers Eroding CTG Effort on project tasks Average commitment per provider Contacts Committed providers Total CTG effort Frac tion of CTG effort on c ollaboration Weight on responsibility Effec t of responsibility of collaboration on contacts CTG Effort on responsibility of collaboration Gaining commitment Saturation effect Lossing commitment CTG effort on Collaboration Time for commitment to break down Positive word of mouth effec t Frac tion of providers committed Provider total population Effec t of responsibility of collaboration on time to commitment to break down Behavior Project 800 2 100 200 4 People*Hour/Month Dmnl People Component Unsolved problem/Component 1 5 5 5 1 4 1 1 34 23 4 4 23 23 4 2 3 5 5 1 0 0 0 0 2 People*Hour/Month Dmnl People Component Unsolved problem/Component 2 1 0 2 4 8 12 16 20 Time (Month) 24 28 Av ailable St at e Effort : Base People*Hour/Mont h 1 1 1 1 Collaborat ion : Base 2 2 2 2 2 2 2 Dmnl Commit t ed providers : Base 3 People 3 3 3 3 3 Feasible Component s : Base 4 4 4 4 4 4 Component Av erage unsolved problems per component : Base 5 5 Un solved problem/Component Second Model Reflector Feedback Conceptual Model Time to solve problem Perc eived risk Unsolved problems Problem per component Problem density Components Ability to work together Effort on facilitation Effort on problem solving Actor 1 Effort Perc eived progress Perc eption of problem density Effort on Making progress Perc eived benefit Actor 1 Trust Collaboration 1 Effort of Ac tors in Problem Solving Activities Agregatte Unsolved problems per component Maximum Unresolved Problems per Component Effort of Ac tor 1 in Problem Solving Activities Generating problems Normalized Problems per Component Initial Unsolved Problems Effec t of Problem Density on Problem Solving Productivity Unsolved Problems Solving problems PSP f Problem Solving productivity Maximum Problem Solving Produc tivity Effort of Ac tor 2 in Problem Solving Activities Actor 1 Effort on project Actor 2 Effort on project Problems per Component (density) Normal Component Building Productivity Component Building Productivity Building ETPCBP f Effec t of time pressure on CB productivity Effort of Ac tors in Producing Components Activities Agreggate EADA1 f Effort Allocation Distribution from Actor 1 Projec t definition Frac tional perc eived progress Frac tion remaining Initial Components Components Effort of Ac tor 2 in Producing Components Activities Effort of Ac tor 1 in Producing Components Activities EADA2 f Effort Allocation Distribution from Actor 2 FINAL TIME Normal produc tivity in problem solving Normal problems per component Problem per component PPC f Creating problems Unsolved problems Solving problems Productivity in problem solving Collaboration 2 EPDP f Effec t of problem density on productivity Problem density TSP f PD f Productivity on creating components Components Creating components Perfec t ability EOPL f Perc eived Progress Perc eived Benefit Time to solve problem Perc eived ris k Time to perceive Trust on Actor 2 Level of perfec tion Time to erode ability Ability to work with Actor 2 Effec t of experience EEL f Effec t of perfection on learning Eroding ability Effort on Making progress Building ability PB f Learning per hour of interaction wt Effort on problem solving wbr Level of engagement Frac tion of effort to task Actor 1 Effort Time to adjust Effort adjustment Indicated Effort Available effort Behavior Main Stocks 8,000 1 400 800 Components Ability Problems People*Hours/Month 0 0 0 0 Components Ability Problems People*Hours/Month 0 5 10 15 20 25 30 35 Time (Month) 40 45 50 Component s : Base Abilit y t o work wit h Act or 2 : Base Un solved problems : Base Act or 1 Effort : Base Component s Abilit y Problems People*Hours/Mont h Problems • Trust1 has no trust • An infectious theory of collaboration? • Collaboration 1 has no collaboration • Collaboration 2 is a single-actor collaboration model (a half-collaboration model?) • Conceptual blurring Critical Theoretical Problems • 1-party, 2-party, multi-party focus • Multiple stages of IT development and scale-up issues – Understanding / specification discovery – Prototype construction – Production system implementation • Dependence on highly abstract variables – What drives changes to TRUST and ENGAGEMENT? What do they do? Where We Are: Approaching Solutions Where We Are (*) We are about here * Diagram taken from: Randers, J., Ed. (1980). Elements of the System Dynamics Method. Cambridge MA, Productivity Press. Third Model Building on Black (2002) Research A's accumulated knowledge A's knowledge of A's work A's knowledge of B's work artifact action location timing cross-boundary activity A learning from the activity, w hen it is ACCESSIBLE to A B learning from the activity, w hen it is ACCESSIBLE to B A PARTICIPATING in the activity at the boundary B PARTICIPATING in the activity at the boundary B's accumulated knowledge B's knowledge of B's work B's knowledge of A's work Model Overview State's Knowledge State's Knowledge of State's Role in the Project State's Knowledge of Provider's Role in the Project + Collaboratively Doing Project Work + R Project Work + R Collaboratively Doing Project Work Provider's Knowledge Provider's Knowledge of State's Role in the Project Provider's Knowledge of Provider's Role in the Project + + Learning by Doing Sense of Progress in Project Learning by Doing + + Sense of Progress in Project Tools that Facilitate Knowledge Elicitation and Communication CTG Facilitative Methods and Tools Initial Trust to Commence Work Initial Trust to Commence Work Project Work concreteness work to do doing new work right work really done doing new work wrong transformability doing rework wrong doing rework right learning by doing undiscovered rework recognizing problems known rework Knowledge, Engagement, and Trust weight on trust Provider's engagement collaborating in the project work Provider's trust State's know ledge of State's project w ork Provider's know ledge of State's project w ork State's learning about State's work Provider learning about State's work State's know ledge of Providers's project w ork State's learning about Provider's work Provider's know ledge of Prov ider's Provider's learning project w ork about Provider's work State's trust State's engagement Model Main reinforcing processes accu racy fro m Pro vider 's po in t o f view CTG i + + co ncr eteness + tran sfor mability State's kno wled ge ab ou t Pr ovid er's wo rk + + R6 sen se of p ro gr ess + + err o r co llab o ratively + d oin g wor k i + Pro vider 's tru st L earn ing (no t) to wo rk with you + Pro vider 's en gag emen t R1 Getting en gag ed o r en rag ed + + R2 Getting en gag ed o r en rag ed + State's en gag emen t + L earn ing (no t) t o wor k with me + State's tr ust R5 R3 + + Pro vider 's kn owled g e abo u t Pro vider 's wo rk L earn ing by d oin g (o r n ot) + R4 L earn ing by d oin g (o r n ot) + + State's kn owled g e abo u t State's wor k + R7 + Pro vider 's kn owled g e abo u t State's wor k + + Getting to kn o w each o th er o r r ecipr ocally with ho ldin g in for matio n accu racy fro m State's p o int o f view Behavior Trust and collaboration 2 1.5 1 0.5 0 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 Time (Month) Dmnl Dmnl Dmnl State's trust : HIMS Provider's trust : HIMS collaborating in the project work : HIMS Behavior Knowledge 1 0.75 0.5 0.25 0 0 8 16 24 32 40 Time (Month) 48 56 64 72 Dmnl Dmnl Dmnl Dmnl State's knowledge of Providers's project work : HIMS State's knowledge of State's project work : HIMS Provider's knowledge of Provider's project work : HIMS Provider's knowledge of State's project work : HIMS Where We Might Go: New Model Simulations Contributions Potential Contributions • Theoretical – Explore scale-up issues in phases of IT work – Explore differences and similarities in interagency and intergovernmental IT work • Practical – Develop a cross-project comparison tool – Use model scenarios as training The 3 Models (summary) • Trust 1 – Centered in Project Dynamics and how these influence trust and collaboration. • Collaboration 2 – Centered around the interaction dynamics of HIMS team members and how these influence trust and collaboration. • Knowledge-Based Trust 1 – Centered around the dynamics generated by the use of facilitative tools and methods in the collaborative effort and how these influence trust and collaboration. The Products Generated • International Conference of the System Dynamics Society – 19th Atlanta 2000 A. A Preliminary System Dynamics Model of Intergovernmental Collaboration B. Group Modeling of IT-Based Innovations in the Public Sector – 20th Palermo 2001 A. Evolution of a Dynamic Theory of Collaboration: Modeling Intergovernmental Use of Information Technology • Hawaiian International Conference on Systems Sciences – HICSS 35 2001 A. Modeling Intergovernmental Collaboration: A System Dynamics Approach – HICSS 36 2002 A. A Dynamic Theory of Collaboration: A Structural Approach to Facilitating Intergovernmental Use of Information Technology