Can I merge IP addresses and MAC addresses if I re-design the network?

I read books related to network basics a while ago. I had a question: Why do I have a MAC address and an IP address? Can the two be combined into one? The current logic is as follows: During data transmission, the vro can view the IP address of the data packet and compare it with the "IP Set: Next Hop MAC address" recorded in the routing table, when the network adapter receives data, it also checks whether the target MAC address recorded in the data packet is itself, and only receives its own MAC address. The MAC address is used by the data link layer, and the IP address is used by the network layer. The MAC address is the physical property of the hardware. It remains unchanged on the NIC, And the IP address is the logical property. Therefore, because of the network layering design, it is natural to think of using two different "addresses ". So is it possible that two layers use the same thing? Can I directly use IP addresses on the data link layer? This is equivalent to a variable MAC address. After all, the acceptor may be a notebook and may run everywhere. Assume that only IP addresses are used.In order to ensure that the data arrives at a computer correctly, the IP address must be bound to a receiver. Once the data changes, the receiver must tell the network that I am a certain IP address, after all, when a route is forwarding data packets, it is also forwarded based on the MAC address bound to the IP address, with a timed refresh mechanism. In this way, the route table contains the ing between the IP address set and the target Route IP address to be forwarded in the next step, The IP address mentioned here is not a virtual attribute of the current online world. It has both physical and logical attributes, representing the logical attributes of the network layer and the physical Nic.. There is no ing between MAC addresses and IP addresses. Is it much simpler? Assume that only the MAC address is used. because the value of the MAC address is not associated with its network location, it is better to forward the previous few identical IP addresses to a specific route, therefore, if only the MAC address is used, the ing relationship recorded by the router is as follows: Next route MAC address: the MAC address of all receivers in the route. When the route is forwarded, you need to traverse and compare all the acceptor MAC addresses to determine the next step. Obviously, the logic becomes very complex and the performance is also very low. If you are in a Small Lan, you can only use the MAC address, but it will not work if it involves a wide network. If the IP address and MAC address are combined into one, the MAC address on the NIC is a dynamically changing value, which comes from the IP address allocated by the carrier. Is there a problem? At present, my understanding should be acceptable. I don't know what the experts think? I have seen that many answers on the network have preset precondition errors. For example, if a book describes that a MAC address is required, the "host uses MAC Address Identification" is first preset ", the MAC address does not exist if the host is identified by an IP address. The above is just a mess. The combination of IP and Mac may be due to historical reasons. IP addresses can unify all kinds of physical networks, while Mac may be only one type of physical network. Layering is advantageous. There is a famous saying that any difficult problem in the computer field can be solved by adding a hierarchy. It is suitable for use here.