<p dir="ltr">Structural instability, traditionally deemed undesirable in engineering, can be leveraged for beneficial outcomes through intelligent design. One notable instance is elastic buckling, often leading to structures with two stable equilibria (bistable). Connecting bistable elements to form multistable mechanical metamaterials can enable the discretization and offer tunability of mechanical properties without the need for continuous energy input.<i> </i>In this work, we study the physics of these multistable metamaterials and utilize their state and property alterations along with snap-through instabilities resulting from state change for engineering applications. These materials hold potential for diverse applications, including mechanical and thermo-mechanical defrosting, energy absorption, energy harvesting, and mechanical storage and computation.</p><p dir="ltr">Focusing on defrosting, we find that the energy-efficient mechanical method using embedded bistable structures in heat exchanger fins significantly outperforms the thermal methods. The combination of manufacturing methods, material choice, boundary conditions, and actuation methodologies is systematically investigated to enhance defrosting performance. A purely mechanical strategy is effective against solid, glaze-like ice accumulations; however, performance is substantially diminished for low-density frost. To address this limitation, we study frost formation on the angular shape morphing fins and subsequently introduce a thermo-mechanical defrosting strategy. This hybrid approach focuses on the partial phase transition of low-density frost to solid ice through thermal methods, followed by mechanical defrosting. We experimentally validate this approach on a multistable heat exchanger fin pack.</p><p dir="ltr">Recent advancements have led to a new paradigm of reusable energy-absorbing materials, known as Phase Transforming Cellular Materials (PXCM) that utilize multiple negative stiffness elements connected in series. We explore the feasibility of this multistable metamaterial as frequency up-conversion material and utilize these phase transformations for energy harvesting. We experimentally demonstrate the energy-harvesting capabilities of a phase-transforming unit-cell-spring configuration and investigate the potential of multicell PXCM as an energy harvesting material.</p><p dir="ltr">The evolution towards intelligent matter, or physical intelligence, in the context of mechanical metamaterials can be characterized into four distinct stages: static, responsive, adaptive, and intelligent mechanical metamaterials. In the pursuit of designing intelligent mechanical metamaterials, there has been a resurgence in the field of mechanical computing. We utilize multistable metamaterials to develop mechanical storage systems that encode memory via bistable state changes and decode it through a global stiffness readout. We establish upper bounds for maximum memory capacity in elastic bit blocks and propose an optimal stiffness distribution for unique and identifiable global states. Through both parallel and series configurations, we realize various logic gates, thereby enabling in-memory computation. We further extend this framework by incorporating viscoelastic mechano-bits, which mimic the decay of neuronal action potentials. This allows for temporal stiffness modulation and results in increased memory storage via non-abelian behavior, for which we define a fundamental time limit of detectability. Additionally, we investigate information entropy in both elastic and viscoelastic systems, showing that temporal neural coding schemes can extend the system’s entropy beyond conventional limits. This is experimentally validated and shown to not only enhance memory storage but also augment computational capabilities.</p><p dir="ltr">The work in this thesis establishes multistability as a key design principle for developing responsive, adaptive, and intelligent materials, opening new avenues for future research in the field of multistable metamaterials.</p>