In a production environment, physical and cyber-related failures become unavoidable because of the complexity of highly connected manufacturing systems and a finite equipment life cycle. The main purpose of this thesis is to investigate optimal maintenance outsourcing and replacement policies in the presence of cyber-threats, as well as policies to achieve channel coordination via cost subsidization. Although well-developed maintenance outsourcing literature has addressed many advanced and trending issues such as the costs and benefits of new technology adoptions, learning effects, forgetting effects, and systems with back-up machines, no study has looked at the effect of cyber threats on connected production systems. Besides filling this gap, this thesis addresses the most common replacement policies including preventive maintenance with minimal repairs and age replacement. The advent of the Internet of Things (IoT) has enabled the creation of "smart" manufacturing systems, However the resulting connected format makes these systems potential targets for cyber-attacks. Manufacturers have to face a difficult decision as to whether they should adopt costly security technologies or let the manufacturing systems be vulnerable to cyber-threats. This study develops a model addressing this dilemma by providing some insights into the effects of cost subsidization and installation of security systems on the system profit. Our findings suggest that a manufacturer can trade off cost subsidization to gain flexibility in the maintenance decision-making process. The study also provides evidence that, for a "smart" manufacturing system, the existence of a security system guarantees at least a marginal increase in system profit. With the help of an extensive numerical study, we provide guidelines for optimizing system profit and addressing the effect of various model parameters (such as a monetary parameter, failure rate and a cyber-attack related parameter) on the value of a security system.In the second part of this thesis, we build analytical models to deal with cyber-threats in the context of a production system that utilizes the age replacement policy. We propose two different scenarios involving cyber-threats. First, cyber-attacks do not disrupt the physical production system. In this case, we replace the broken components either when the system has a physical breakdown or when a predetermined amount of time (T) elapses. Second, we need to replace the system right after it suffers from a successful cyber-attack. Thus, we replace the system components if one of three following events happen, whichever comes first: (1) a successful cyber-attack, (2) a physical breakdown, or (3) a predetermined time (T) elapses. We analytically provide optimal policies to minimize the expected cost for these scenarios. Finally, we carry out sensitivity analyses to show the effect of different model parameters on the value of a cyber security system.