That's an excellent question, as it strikes right at the general strategy of the Gaulish coalition of -52. Indeed, Caesar's armies capacities of projection in Gaul depended of supplies lines, which were fairly reliable thanks to the developed road network in Gaul (it's common knowledge at this point that, especially in Celtica, most Roman roads were following Gaulish pathways) and the support of Gaulish clients, in particular Aiduoi (Aeduii).

Vercingetorix basically attempted to deprive Romans from benefiting from this, thanks to general defection of most Gaulish polities, slaughter of Roman traders and suppliers at Cenabum and Cabillonum, and resorting to guerilla tactics (which worked well enough in Belgica) or scorched earth.In order to apply this military program, Vercingetorix fully used his status of leader of the coalition decided by the pan-Gaulish assembly (an institution that predated the war, and seems to have functioned more or less regularly at least since the late IIIrd century BCE) to impose his plan to various Gaulish peoples and tribes (very roughly, Gaulish polities gathered several teutas, tribes) : pro-roman factions were loosing ground and even Aedui, who supplied Caesar with grain, joined the coalition, as their own anti-roman factions took over (with difficulty and because they wanted to prevent Arverns to gain the political primacy in Gaul).

Caesar, who was in Italy and having supplied armies to oppose Vercingetorix's managed to quickly counter-attack in the first months of -52 (especially thanks to mobile units such as the German cavalry), but didn't managed to decisively defeat pan-Gaulish forces, and had to retreat to Transalpine Gaul (arguably preventing Vercingetorix to attack his moving army) to resupply, then coming back during the Spring to attack Avaricum.The city and its reserves were spared by Vercingetorix, either because local elites persuaded him not too ruin one of Gaul's wealthiest places, either because it was Vercingetorix plan to force Caesar to besiege the place, attracting him with the promises of looted supplies.Considering the overall strategy of the coalition, this second option is plausible, searching to force Caesar to a long siege and trough harassing tactics and provocation, exploit an attrition of the Roman army, which failed due to Caesar's tactics and successful assault : regardless of what was planned, it put Vercingetorix in a difficult position, while he kept the leadership, being accused of pusillanimity or cowardice.

The Siege of Gergovia represented something similar, with Vercingetorix "fixing" Caesar's army who seems to have began to build a complex network of fortifications, which were left unfinished : overall, the meta-capitol of Arverns (which was made of three oppidae close to each other : Gondole/Corent/Gergovia) didn't let much room for an easy siege and Caesar almost made a fatal mistake by electing to assault Vercingetorix; as circumvallating the oppidae was not an option. On the other hand, while a clear military and moral victory for Gauls, Vercingetorix didn't really inflicted any real damage to Romans or deprived them of their supplies or remaining arsenals in Gaul, while attempts at stirring up Gauls in Transalpina didn't managed to really make a dent at Roman capacities there.

When Caesar managed to join up with Labienus (who defeated Gauls at Lutetia), he reinforced his army with not only Romans, Germanic auxiliaries and Gaulish allies, but as well with new supplies gathered previously and safeguarded. When he move back to Transalpina, Caesar doesn't really have a logistical problem in spite of Gaulish efforts who, on the contrary, had requisitioned what they could in a country that years of war damaged (most of forces coming from western Gaul were probably significantly reduced).It's maybe, realizing the situation and fearing to see his authority put in question if he didn't act, why Vercingetorix rushed to harass Caesar, electing to use Alisia/Alesia, a strongly fortified place on the road Caesar had to take to Transalpina, as a fixing point like Gaulish attempts at Avaricum and Gergovia.A first harassing attack mostly failed to have any lasting effect other than forcing Caesar to follow Vercingetorix' army (80 000 to 100 000 men, an army he couldn't just ignore) while remaining Gaulish cavalrymen ravaged the countryside to deprive Caesar further supplies. But Caesar had enough supplies gathered in Gaul during the Spring, to hold for a while, and further provocations from Gaul failed to force Caesar to a suicidal attack against their fortifications. Instead, Caesar was in a much more better situation to counter-fortify the place AND against the incoming rear army.

Vercingetorix' plan depended from a successful harassment of Roman forces coordinated attacks with autonomous Gaulish commanders, and a quick military resolution : but superior Roman tactics (while Gaulish armies eventually, by -53/-52, managed to counter Roman tactics, there was no time or capacity to systematize it, and Gaulish poliorcetics were fairly limited), inability to prevent Roman to re-supply their forces or their reinforcement after the Battle of Lutetia, and being stuck with ten of thousands men without a real tradition of large-scale logistics did the rest.Eventually, in this situation, time favoured Caesar, who knew that as long he prevented the second Gaulish army to breach down his double fortification (which was admittedly close enough to have happened), he could obtain a decisive and crushing victory.

- De Bello Gallico, from some Mr Caesar, Jules
- Vercingétorix; Jean-Louis Brunaux; 2018
- Vercingetorix, chef de guerre; Alain Deyber; 2017