Thesis (Ph. D.)--University of Wisconsin--Madison, 1976. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 143-148).

No description available.

No description available.

No description available.

published_or_final_version / Computer Science / Master / Master of Philosophy

No description available.

No description available.

No description available.

Multiteam systems (MTSs), defined as two or more interdependent teams working towards both proximal team goals and at least one shared goal, are prevalent in modern organizations. Prior research has shown that MTS effectiveness is a function of the quality of both the processes occurring within each component team and between the teams in the system (Marks, DeChurch, Mathieu, Panzer, & Alonso, 2005; DeChurch & Marks, 2006). The critical drivers of both team and MTS effectiveness include behavioral processes (explicit actions directed towards others; e.g., communication), cognitive states (knowledge or perceptions; e.g., transactive memory), and affective states (emotions or mood; e.g., stress) emerging from the shared experiences of the members of the team (Cohen & Bailey, 1997; Marks, Mathieu, & Zaccaro, 2001; Mathieu, Marks, & Zaccaro, 2001). While these phenomena exist both within and between teams, prior research has shown that such processes and states cannot be assumed equivalent across these levels (DeChurch & Zaccaro, 2010). Further complicating these relationships, these processes and states are expected to impact the relationships that other phenomena have on performance in addition to their expected direct effects (Ilgen, Hollenbeck, Johnson, & Jundt, 2005). With this, the purpose of this thesis is to study the relationships between process, cognitive and affective states, and performance as each exists within and between teams. Central to this purpose is examining the effects of cognitive and affective states on the relationship between process and performance. These relationships were tested using a laboratory sample of six-person MTSs (N = 118, n = 708) performing an action- and information sharing-oriented task. Utilizing network analysis, the direct and conditional impact of behavioral process (i.e., communication), cognitive states (i.e., advice relationships), and affective states (i.e., hindrance relationships) within and between teams were captured. It was found that the impact of between-team communication on MTS performance was moderated by between-team advice relationships and the impact of within-team communication on team performance was moderated by within-team hindrance relationships. Together, these findings suggest a need to consider the effects of within- and between-team processes on performance as having different conditional relationships with co-occurring states.

With the emergence of advanced integrated services networks, the need for effective performance analysis techniques has become extremely important. Further advancements in these networks can only be possible if the practical performance issues of the existing networks are clearly understood. This thesis is concerned with the design and development of a measurement system which has been implemented on a large experimental network. The measurement system is based on dedicated traffic generators which have been designed and implemented on the Project Unison network. The Unison project is a multisite networking experiment for conducting research into the interconnection and interworking of local area network based multi-media application systems. The traffic generators were first developed for the Cambridge Ring based Unison network. Once their usefulness and effectiveness was proven, high performance traffic generators using transputer technology were built for the Cambridge Fast Ring based Unison network. The measurement system is capable of measuring the conventional performance parameters such as throughput and packet delay, and is able to characterise the operational performance of network bridging components under various loading conditions. In particular, the measurement system has been used in a 'measure and tune' fashion in order to improve the performance of a complex bridging device. Accurate measurement of packet delay in wide area networks is a recognised problem. The problem is associated with the synchronisation of the clocks between the distant machines. A chronological timestamping technique has been introduced in which the clocks are synchronised using a broadcast synchronisation technique. Rugby time clock receivers have been interfaced to each generator for the purpose of synchronisation. In order to design network applications, an accurate knowledge of the expected network performance under different loading conditions is essential. Using the measurement system, this has been achieved by examining the network characteristics at the network/user interface. Also, the generators are capable of emulating a variety of application traffic which can be injected into the network along with the traffic from real applications, thus enabling user oriented performance parameters to be evaluated in a mixed traffic environment. A number of performance measurement experiments have been conducted using the measurement system. Experimental results obtained from the Unison network serve to emphasise the power and effectiveness of the measurement methodology.