There are a couple of interesting things here. The first is the use of output devices used on this board - they are "logic Level FETs" - 20N10L - when I went to grab the data sheet this is what I found:

"STP20N10L - OLD PRODUCT: NOT SUITABLE FOR NEW DESIGN-IN    ST Microelectronics"

So this confirms my idea of steering clear of some of these 'new fangled' parts - they go out of fashion way too quickly and then you have a useless board????

The 47K resistors to ground (looks like 47K in the bottom pic) are used to ensure the LLFs are held in the OFF state as they are driven by a 5V logic rail - IE directly from the 74240.

74240 are '3 state output devices' BUT they are NOT 'open collector' outputs that would require a tie high resistor or load. The outputs are configured by the designer to be one of the 'tri' states as she needs it - either on the fly or certainly at boot.

The board should work OK as is and I suspect the 'added' 47K resistors were needed because as the 74240 was being initialised to force its tri-state outputs low (at boot probably) these resistors ensure the driver transistors stay OFF and not fire the solenoids. This reminds me so much of the RUBBISH Gottlieb SS design - where all solenoids fire when you power up the machine..….design with no thought!

If I ever design something this crap please come and burn my workshop down!