The　optimization　of　a　packed　bed　for　utility-scale　applications　is　presented　in　this　paper.　The　effects　that　particle　size,　aspect　ratio　and　storage　mass　have　on　the　roundtrip　exergy　efficiency　of　the　store　are　thoroughly　analysed.　The　paper　seeks　to　provide　a　clear　insight　of　what　ranges　of　values　for　the　aforementioned　design　parameters　are　adequate　to　consider　when　designing　a　grid-scale　packed　bed.　Simulations　were　carried　out　using　a　one-dimensional　model　that　accounts　for　temperature-dependent　properties　and　self-discharge　losses.　The　assumed　operating　temperature　range　for　the　packed　bed　is　290-823　K,　which　is　typical　of　CSP　plants　and　CAES　systems.　A　24-h　work　cycle　(12　h　charge/12　h　discharge)　with　variable　power　(10　MW　peak)　and　a　total　energy　storage　requirement　of　79.4　MWh(th)　has　been　considered　for　the　study.　It　has　been　found　that　exergy　losses　are　minimized　if　a　configuration　based　on　an　aspect　ratio　between　0.5　and　0.8　is　adopted　and　the　size　of　the　rocks　is　finely　tuned　for　the　specific　shape　of　container.　In　this　work-unlike　similar　studies-a　cost-benefit　analysis　has　been　carried　out,　which　indicates　that　increasing　the　thermal　storage　mass　leads　to　a　considerable　increase　in　efficiency.　A　mass　overrating　of　50%　yields　the　lowest　levelized　cost　of　storage　for　the　economic　scenario　considered.　The　optimum　design　obtained　from　the　optimization　process　has　an　aspect　ratio　of　0.6,　a　particle　size　of　4　mm　and　a　mass　overrating　factor　of　1.5.　This　packed　bed　attained　a　roundtrip　exergy　efficiency　of　98.24%.