Hybrid project control system design for product development projects based on lean principles

For this dissertation project, a new project control system is designed for product development projects. Proposed control system and tools are aligned with lean principles. Pull and push frameworks are integrated to form an hybrid control system. Product development begins with the realisation of the opportunity, ends with manufacturing of goods or services. Product development process time and results are variable and process priority relations are complex as product development mainly consists of problem solving and information processing. In addition to that, product development requires a group of experts from diverse fields to work closely for a long time to achieve a specific goal. Despite all this difficulties, improvement efforts focusing on product development systems promises high potential gains for several reasons; (i) product development is the first stage to identify customer value definition and (ii) product development decisions affects all functions of an organization for a period of product lifecycle. This study aims to design a new project control system that uses a team based organization structure to gather wide range of expertise to control project flow through diverse processes. Elements of pull control is used to create a continuous development environment and to absorb the detrimental effects of high variability and overburden. Push control is integrated with pull control to form an hybrid system due to its flexibility. The objective of the proposed control system is to create flow and reduce product development lead time. The fundamentals of lean thinking is laid in the '90s and since then it has been applied in many fields such manufacturing, logistics, health care, accounting. Lean thinking, classifies processes as value-added and non-value added (waste) based on customer's value definition and aims to eliminate non-value added steps to create uninterrupted flow of goods or information. Lean tools facilitate waste identification and elimination through control of the root causes of waste, variability and overburden. Lean tools also helps to visualize the flow to detect non value added activities. To apply lean thinking to product development processes, conventional lean tools, such as value stream mapping and kanban are used in accordance with tools specific to product development such as design structure matrix and Oobeya. Waste categories that are used for any other application area, can also be applied to product development systems. Lean product development principles are also derived in line with original lean principles. Waste and redefined the concept of value in the product development system. Lean product development principles to the product development systems compatible with Lean principles have also been defined. The proposed project control system is based on lean principles and tools of lean thinking are different interactive with the thesis has been put together in such a way. For the proposed hybrid project control system, different lean tools are integrated to design a system that conforms lean principles. Literature review is performed on two levels for both project scheduling literature and lean product development principles. An increasing trend is observed for project scheduling literature over the last thirty years. A great number of studies has been published on the topic of multi project scheduling. Yet, project scheduling articles that focuses on product development is limited. There are number of studies that addresses both lean thinking and product development. But literature search revealed no results for an article that covers both project scheduling and lean. Also detailed literature search on project management and scheduling shows the gap for the proposed hybrid project control system. Literature search for lean product development principles show that a majority of the literature focuses on value definition and waste identification, visual information sharing issues, whereas balancing cross functional expertise and sustaining continuous improvement culture issues are neglected. Proposed methodology for hybrid project control system design for product development projects consists of six stages in multi layered, hierarchical structure. Stages are further divided in steps and substeps to provide a detailed road map for practitioners and researchers. The methodology starts with current system analysis on the first stage. Data concerning process definitions, durations and resource requirements are collected. System analysis results are visualized by value stream mapping and design structure matrix tools. Value stream mapping is a generic visualization tool for lean implementation. But for the proposed system it is further advanced to capture system specific attributes. Design structure matrix conveys process relation information in a concise format to later alter project dependencies. Second stage of the methodology involves in-depth analysis of reverse flows and iterations of product development flow. Design structure matrix is first resequenced, then clustered to form subproject with minimum reverse flow. Current labour assignment performance is analysed on stage three and an alternative team based assignment is performed in accordance to takt time. At this stage, competency matrix are used to analyse and plan for future requirements if the resulting team based assignment may require competency advancement of current workforce. The fourth stage is the focal point of the proposed system, that is pull-push hybrid control. Different prioritization rules are provided to better align push control to selected application area and kanban board examples are provided for pull control system design. Action plan with durations and future state value stream map forms fifth stage directly serves the implementation stage. Last stage is the implementation and later, following a predetermined amount of time, methodology recommence in a cyclic fashion for continuous improvement. Two real life industrial case studies are used to evaluate the proposed system. Case study companies are selected to cover different types product characteristics, organisational structure and scale to explore the limits of applicability of the proposed system. The first case study is set on office furniture development system. Product development efforts from product definition to manufacturing is analysed. Value stream mapping and design structure matrix is formed on the first stage of the methodology to visualize system analysis results. Based on takt time information, processes are later grouped to form sub projects by clustering design structure matrix. Several alternative design structure matrix sequences and clustering results are compared and evaluated. Current labour assignment is analysed and a team based assignment is realised for process groups. Required competency expansion is planned and visualized by using competency matrix. For the push section of the control system, different rules are evaluated and detailed kanban boards are designed for pull control. Future state value stream map and action plan with calendar information is provided for implementation. For this case study a simulation model is also used to evaluate several key characteristics for project control system design that is variability and utilization. Results shows that the use of proposed project control system results in better lead time performance. The second case study is set on white-goods sector. System scale and hierarchic nature of the system is reflected on the first stage of the methodology. First overall product development system is analysed. Both current state value stream map and design structure matrix are created. Then a part of the flow is selected and later analysed in detail by creating a more detail current state map and design structure matrix for that specific section. All stages of the proposed methodology is performed for verification testing section of the flow. Results of different clustering techniques are evaluated and further examined to fit the real life constraints. Selected clustering structure is used to design testing cells and labour assignment is performed based on takt time. Competency matrix are formed and used to plan for competency expansion for proposed cell structure. An original kanban board is designed, utilization rules are set and a framework for push segment of the control system is developed. On the last stage a detail action plan is provided together with future state value stream map. Case studies verified the performance of the proposed system on different settings. The proposed hybrid project control system is adaptable and instructions for each stage of the methodology is provided for further dissemination to different product development systems. Also selected parts of the methodology can also be applied in ısolation for targeted improvement efforts. This study promises several key areas of further research. Future studies will focus on different pull system design and project sequencing problem.