The 'concept' is good but I'm afraid that's about where it stops. As you can probably guess I have done a LOT of work towards this. Let me point out a few of the things I have stumbled at...some of these might seem small things but they can break a project. As you are proposing a private project not for commercial sale I won't mention licencing or copyright issues as they wouldn't apply.
Playfield - It has to be made from the correct thickness and grade of ply (that simply isn't available in Australia) OR you will run into hardware fitting issues and several other problems.
Theme - you will NEVER get 2 people to agree on a theme let alone a small group who are going to put a lot of time and money into the project. The theme MUST be chosen and stuck to. It is impossible to suggest that your machine will be "Cops & Robbers" and Billys' machine will be "Fireball". There is far too much that relies on the theme. Artwork (cabinet, backglass, playfield, plastics), Electronics (sounds, gameplay [programming]), etc etc.
Electronics - no problem with this in a 'general' sense however, the economies of scale work against you with a project like this. The most cost effective run size is usually a run of 100pcs BUT most components are supplied in bags, usually of several thousand. This isn't a 'deal breaker' but it needs to be kept in mind.
There are many many many many (did I mention MANY?) other things that will cause grief and cause the project to stall. Billys' wife will get the shits after 3 months and demand he pulls out of the project leaving someone else with a screaming baby.........etc etc etc
OK - that's some of the negative stuff. Some positive.
I have spent a LOT of time working through this concept and have settled on a 'way' of presenting to people like yourself the building blocks for the electronics to make a machine happen on a 'build one machine yourself' basis.
There are a LOT of compromises that you will have to accept up front or the project will stall. One is the playfield. The ONLY cost effective way is to make it from MDF and cover with a digitally printed graphic for the playfield. Purists will not like this BUT I point out that back in the day when SS machines were introduced I would spend hours covering the playfields of freshly unpacked machines (Playboy, Kiss etc) with clear contact - so what's the difference??? None really!
Forget about running ANY project from a PC (apologies to Coconut Island) - PCs are unreliable, change every five minutes, become obsolete and are generally a real pain in the arse! Look at the headaches people have getting virtual pins to run ar pretty well ANY MAME setup you have ever seen - forget PCs. The ONLY way is embedded, just like EVERY commercial pin ever made.
I am working on a series of boards that are 'stand alone'. For example, my pop bumper board takes the switch signal from a PB assembly and controls it 100%. It knows when to drive the coil, for how long (adjustable by you), sends a signal to the main MPU for scoring, sound generation etc and has an output to fire a flash/LED/lamp etc. It also knows if the switch has stuck and disables the coil so it can't burn out and reports a fault via the onboard LED.
Add to this separate, small autonomous boards that control kickers, drop targets, ramp flaps and any other feature you wish to incorporate - all talking back to the main (general purpose) MPU and you have a machine - well, electronically anyway.
It eliminates all switch and lamp matrix issues as there are none! It simplifies construction and faultfinding as any fault is contained to the sub-module.
I have PB, drop target and kicker modules working right now.
That is the direction I am heading at the moment......input welcome.