Contents lists available at ScienceDirect Engineering Structures journal homepage: www.elsevier.com/locate/engstruct An investigation of AASHTO’s requirements for providing continuity in simple span bridges made continuous Fatmir Menkulasi a, ⁎ , Al Patel b , Hadi Baghi a a Department of Civil and Environmental Engineering, Wayne State University, Detroit, MI 48202, USA b Clark Nexsen, Virginia Beach, VA 23462, USA ARTICLE INFO Keywords: Time dependent analysis Restraint moments Continuity age Creep Shrinkage Temperature gradients ABSTRACT AASHTO’s requirements for providing continuity in simple span bridges made continuous are investigated by performing a parametric study which consists of 140 time dependent analyses for various precast concrete beam shapes. The beam shapes considered include three precast concrete bulb tees (PCBT), two AASHTO type beams, and two Florida I-beams (FIB). Various beam spacing and span configurations are considered. A sectional analysis approach that employs the age adjusted effective modulus method for capturing creep effects is pre- sented for calculating restraint moments. AASHTO models for creep and shrinkage are used to conduct the parametric study. For each case considered a minimum girder age for when continuity can be established is recommended such that the total restraint moment at the intermediate support is equal to or smaller than zero. The recommended minimum girder ages at continuity vary from 55 to 90 days for PCBTs, 55–70 days for AASHTO type beams, and 55–80 days for FIBs. The influence of ′ f ci , choice of creep and shrinkage model, and choice of analysis method on the magnitude of restraint moments is investigated. The specification of a higher ′ f ci is an effective technique to reduce the minimum required girder age at continuity. The magnitude of restraint moments appears to be highly sensitive to the selected creep and shrinkage model. The proposed analysis method addresses the shortcomings of other closed form formulations and results in restraint moments that are more sensitive to the girder age at continuity. 1. Introduction The advantages of creating continuity in bridges composed of pre- cast beams have been embraced by the engineering community in the United States since the 1960s [1,2]. Continuous bridges provide re- dundancy for overload conditions and extreme events such as vehicular impact, blasts, storm surges or an earthquake. Additionally, the con- tinuity improves rideability and increases the durability of the bridge by eliminating joints at beam ends. It also increases the structural ef- ficiency of the bridge superstructure by making possible longer spans and greater beam spacing. Accordingly, it is essential to ensure that precast beam bridges designed as continuous for live loads do indeed behave as intended. AASHTO LRFD Specifications [3], require that such continuous bridges be designed for restraint moments developed due to time dependent effects or other deformations. These restraint moments could be caused by creep, shrinkage, temperature gradients and support settlements (Fig. 1). The restraint moments can be positive or negative and are typically computed at interior supports of continuous bridges, albeit they affect the design moments at all locations along the bridge. The magnitude and direction of the restraint moments depend on the beam age at the time continuity is established, properties of the beam and deck concrete, and bridge and beam geometry [4]. The commentary of AASHTO LRFD Specifications [3] Article C15.14.1.4.2 states that the data show that the later the continuity is formed, the lower the predicted values of positive restraint moment. Accordingly, it is considered beneficial to wait as long as possible after the beams are cast to establish continuity and cast the deck. Although an early age of continuity can lead to positive restraint moments that may cause cracking at the bottom of continuity diaphragm and affect the efficiency of the continuity at the interior supports, a late age of continuity will maximize negative restraint moments due to differential shrinkage, which could cause transverse cracking on the top of the deck. However, since data form various projects [5,6] do not show the effects of differential shrinkage, it is questionable whether negative moments due to differential shrinkage form to the extent predicted by analysis [3]. Newhouse et al. [7] compared measured restraint mo- ments developed during the early ages of continuity to predicted values obtained using a computer program RMCalc [8,9], and concluded that https://doi.org/10.1016/j.engstruct.2017.12.019 Received 26 June 2017; Received in revised form 27 November 2017; Accepted 12 December 2017 ⁎ Corresponding author. E-mail addresses: fatmir.menkulasi@wayne.edu (F. Menkulasi), apatel@clarknexsen.com (A. Patel), gk0747@wayne.edu (H. Baghi). Engineering Structures 158 (2018) 175–198 Available online 04 January 2018 0141-0296/ © 2017 Elsevier Ltd. All rights reserved. T